add models that have input in the prompt

with_input/1a_52k/best_model/adapter_config.json

@@ -0,0 +1,21 @@
+{
+  "base_model_name_or_path": "huggyllama/llama-7b",
+  "bias": "none",
+  "enable_lora": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "lora_alpha": 16,
+  "lora_dropout": 0.05,
+  "merge_weights": false,
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 16,
+  "target_modules": [
+    "q_proj",
+    "k_proj",
+    "v_proj",
+    "o_proj"
+  ],
+  "task_type": "CAUSAL_LM"
+}

with_input/1a_52k/best_model/adapter_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c2b1bd67183f9a1baf20d70bc2f133c2eb83414e83a92f974a08cdeeab1e20bc
+size 67201357

with_input/1a_52k/best_model/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7125032c106c1d32fcd5e1faa58077c673f5fa8158602e6ce96b3a14abc9d730
+size 134433093

with_input/1a_52k/best_model/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2a434d34ef784fd69d9dbf4ca0786e79c78ddb09afd0bd112cf41afcd495d2bc
+size 14575

with_input/1a_52k/best_model/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:783ca115c5e885872176fe1c7b9f675c423d52e34b3401bb0a09dc5a7c1ea241
+size 627

with_input/1a_52k/best_model/trainer_state.json

@@ -0,0 +1,1939 @@
+{
+  "best_metric": 0.6470396518707275,
+  "best_model_checkpoint": "checkpoints/instrucode/with_input/1a_52k/checkpoint-3000",
+  "epoch": 1.9198464122870171,
+  "eval_steps": 200,
+  "global_step": 3000,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.6999999999999996e-05,
+      "loss": 1.8998,
+      "step": 10
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.6999999999999996e-05,
+      "loss": 1.8477,
+      "step": 20
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 8.4e-05,
+      "loss": 1.5682,
+      "step": 30
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 0.00011399999999999999,
+      "loss": 1.1025,
+      "step": 40
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 0.00014399999999999998,
+      "loss": 0.9534,
+      "step": 50
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 0.00017399999999999997,
+      "loss": 0.9024,
+      "step": 60
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 0.000204,
+      "loss": 0.7691,
+      "step": 70
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 0.000234,
+      "loss": 0.712,
+      "step": 80
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 0.00026399999999999997,
+      "loss": 0.7302,
+      "step": 90
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 0.000294,
+      "loss": 0.6993,
+      "step": 100
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 0.0002996887159533074,
+      "loss": 0.6977,
+      "step": 110
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 0.0002992996108949416,
+      "loss": 0.6908,
+      "step": 120
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 0.00029891050583657586,
+      "loss": 0.6971,
+      "step": 130
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 0.0002985214007782101,
+      "loss": 0.6838,
+      "step": 140
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 0.00029813229571984433,
+      "loss": 0.6588,
+      "step": 150
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 0.0002977431906614786,
+      "loss": 0.672,
+      "step": 160
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 0.0002973540856031128,
+      "loss": 0.6933,
+      "step": 170
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 0.00029696498054474705,
+      "loss": 0.6926,
+      "step": 180
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 0.0002965758754863813,
+      "loss": 0.6862,
+      "step": 190
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 0.00029618677042801553,
+      "loss": 0.672,
+      "step": 200
+    },
+    {
+      "epoch": 0.13,
+      "eval_loss": 0.6792568564414978,
+      "eval_runtime": 292.2523,
+      "eval_samples_per_second": 6.843,
+      "eval_steps_per_second": 0.855,
+      "step": 200
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 0.00029579766536964977,
+      "loss": 0.6773,
+      "step": 210
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 0.000295408560311284,
+      "loss": 0.652,
+      "step": 220
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 0.00029501945525291825,
+      "loss": 0.662,
+      "step": 230
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 0.0002946303501945525,
+      "loss": 0.6919,
+      "step": 240
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 0.00029424124513618673,
+      "loss": 0.6828,
+      "step": 250
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 0.00029385214007782097,
+      "loss": 0.6734,
+      "step": 260
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 0.0002934630350194552,
+      "loss": 0.6605,
+      "step": 270
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 0.00029307392996108945,
+      "loss": 0.6779,
+      "step": 280
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 0.0002926848249027237,
+      "loss": 0.6751,
+      "step": 290
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 0.00029229571984435793,
+      "loss": 0.6514,
+      "step": 300
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 0.00029190661478599217,
+      "loss": 0.6764,
+      "step": 310
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 0.0002915175097276264,
+      "loss": 0.6946,
+      "step": 320
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 0.00029112840466926065,
+      "loss": 0.6687,
+      "step": 330
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 0.00029073929961089494,
+      "loss": 0.6504,
+      "step": 340
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 0.0002903501945525292,
+      "loss": 0.6598,
+      "step": 350
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 0.0002899610894941634,
+      "loss": 0.6719,
+      "step": 360
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 0.00028957198443579766,
+      "loss": 0.6514,
+      "step": 370
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 0.0002891828793774319,
+      "loss": 0.6696,
+      "step": 380
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 0.00028879377431906614,
+      "loss": 0.6726,
+      "step": 390
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 0.0002884046692607004,
+      "loss": 0.6505,
+      "step": 400
+    },
+    {
+      "epoch": 0.26,
+      "eval_loss": 0.6674855351448059,
+      "eval_runtime": 303.0692,
+      "eval_samples_per_second": 6.599,
+      "eval_steps_per_second": 0.825,
+      "step": 400
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 0.0002880155642023346,
+      "loss": 0.6722,
+      "step": 410
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 0.00028762645914396886,
+      "loss": 0.6618,
+      "step": 420
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 0.0002872373540856031,
+      "loss": 0.6679,
+      "step": 430
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 0.00028684824902723734,
+      "loss": 0.6371,
+      "step": 440
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 0.0002864591439688716,
+      "loss": 0.6574,
+      "step": 450
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 0.0002860700389105058,
+      "loss": 0.6779,
+      "step": 460
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 0.00028568093385214006,
+      "loss": 0.674,
+      "step": 470
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 0.0002852918287937743,
+      "loss": 0.6661,
+      "step": 480
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 0.00028490272373540853,
+      "loss": 0.6515,
+      "step": 490
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 0.0002845136186770428,
+      "loss": 0.6719,
+      "step": 500
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 0.000284124513618677,
+      "loss": 0.6875,
+      "step": 510
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 0.00028373540856031125,
+      "loss": 0.6976,
+      "step": 520
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 0.0002833463035019455,
+      "loss": 0.6636,
+      "step": 530
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 0.00028295719844357973,
+      "loss": 0.6481,
+      "step": 540
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 0.00028256809338521397,
+      "loss": 0.6504,
+      "step": 550
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 0.0002821789883268482,
+      "loss": 0.6776,
+      "step": 560
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 0.00028178988326848245,
+      "loss": 0.6906,
+      "step": 570
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 0.0002814007782101167,
+      "loss": 0.6714,
+      "step": 580
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 0.00028101167315175093,
+      "loss": 0.6495,
+      "step": 590
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 0.00028062256809338517,
+      "loss": 0.6508,
+      "step": 600
+    },
+    {
+      "epoch": 0.38,
+      "eval_loss": 0.6634581089019775,
+      "eval_runtime": 294.3504,
+      "eval_samples_per_second": 6.795,
+      "eval_steps_per_second": 0.849,
+      "step": 600
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 0.0002802334630350194,
+      "loss": 0.6613,
+      "step": 610
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 0.0002798443579766537,
+      "loss": 0.6702,
+      "step": 620
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 0.00027945525291828794,
+      "loss": 0.649,
+      "step": 630
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 0.0002790661478599222,
+      "loss": 0.6718,
+      "step": 640
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 0.0002786770428015564,
+      "loss": 0.6647,
+      "step": 650
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 0.00027828793774319066,
+      "loss": 0.6441,
+      "step": 660
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 0.0002778988326848249,
+      "loss": 0.6628,
+      "step": 670
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 0.00027750972762645914,
+      "loss": 0.6358,
+      "step": 680
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 0.0002771206225680934,
+      "loss": 0.6568,
+      "step": 690
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 0.0002767315175097276,
+      "loss": 0.6428,
+      "step": 700
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 0.0002763424124513618,
+      "loss": 0.6644,
+      "step": 710
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 0.00027595330739299605,
+      "loss": 0.6401,
+      "step": 720
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 0.00027556420233463034,
+      "loss": 0.6733,
+      "step": 730
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 0.0002751750972762646,
+      "loss": 0.6773,
+      "step": 740
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 0.0002747859922178988,
+      "loss": 0.667,
+      "step": 750
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 0.00027439688715953306,
+      "loss": 0.6729,
+      "step": 760
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 0.0002740077821011673,
+      "loss": 0.6609,
+      "step": 770
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 0.00027361867704280154,
+      "loss": 0.6349,
+      "step": 780
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 0.0002732295719844358,
+      "loss": 0.6647,
+      "step": 790
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 0.00027284046692607,
+      "loss": 0.6557,
+      "step": 800
+    },
+    {
+      "epoch": 0.51,
+      "eval_loss": 0.6586535573005676,
+      "eval_runtime": 307.0004,
+      "eval_samples_per_second": 6.515,
+      "eval_steps_per_second": 0.814,
+      "step": 800
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 0.00027245136186770426,
+      "loss": 0.6461,
+      "step": 810
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 0.0002720622568093385,
+      "loss": 0.6451,
+      "step": 820
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 0.00027167315175097274,
+      "loss": 0.6556,
+      "step": 830
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 0.000271284046692607,
+      "loss": 0.6176,
+      "step": 840
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 0.0002708949416342412,
+      "loss": 0.6425,
+      "step": 850
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 0.00027050583657587545,
+      "loss": 0.658,
+      "step": 860
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 0.0002701167315175097,
+      "loss": 0.6498,
+      "step": 870
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 0.00026972762645914393,
+      "loss": 0.6337,
+      "step": 880
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 0.00026933852140077817,
+      "loss": 0.6435,
+      "step": 890
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 0.0002689494163424124,
+      "loss": 0.6735,
+      "step": 900
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 0.0002685603112840467,
+      "loss": 0.6674,
+      "step": 910
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 0.00026817120622568095,
+      "loss": 0.6515,
+      "step": 920
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 0.0002677821011673152,
+      "loss": 0.6725,
+      "step": 930
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 0.0002673929961089494,
+      "loss": 0.6484,
+      "step": 940
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 0.0002670038910505836,
+      "loss": 0.6595,
+      "step": 950
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 0.00026661478599221785,
+      "loss": 0.6581,
+      "step": 960
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 0.0002662256809338521,
+      "loss": 0.7032,
+      "step": 970
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 0.00026583657587548633,
+      "loss": 0.6614,
+      "step": 980
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 0.00026544747081712057,
+      "loss": 0.6715,
+      "step": 990
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 0.0002650583657587548,
+      "loss": 0.6224,
+      "step": 1000
+    },
+    {
+      "epoch": 0.64,
+      "eval_loss": 0.6567265391349792,
+      "eval_runtime": 296.8164,
+      "eval_samples_per_second": 6.738,
+      "eval_steps_per_second": 0.842,
+      "step": 1000
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 0.00026466926070038905,
+      "loss": 0.6769,
+      "step": 1010
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 0.00026428015564202334,
+      "loss": 0.6349,
+      "step": 1020
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 0.0002638910505836576,
+      "loss": 0.6293,
+      "step": 1030
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 0.0002635019455252918,
+      "loss": 0.6704,
+      "step": 1040
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 0.00026311284046692606,
+      "loss": 0.6562,
+      "step": 1050
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 0.0002627237354085603,
+      "loss": 0.6572,
+      "step": 1060
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 0.00026233463035019454,
+      "loss": 0.6496,
+      "step": 1070
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 0.0002619455252918288,
+      "loss": 0.6933,
+      "step": 1080
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 0.000261556420233463,
+      "loss": 0.6777,
+      "step": 1090
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 0.00026116731517509726,
+      "loss": 0.6612,
+      "step": 1100
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 0.0002607782101167315,
+      "loss": 0.6434,
+      "step": 1110
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 0.00026038910505836574,
+      "loss": 0.6605,
+      "step": 1120
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 0.00026,
+      "loss": 0.6629,
+      "step": 1130
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 0.0002596108949416342,
+      "loss": 0.6511,
+      "step": 1140
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 0.00025922178988326846,
+      "loss": 0.6427,
+      "step": 1150
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 0.0002588326848249027,
+      "loss": 0.6788,
+      "step": 1160
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 0.00025844357976653694,
+      "loss": 0.6766,
+      "step": 1170
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 0.0002580544747081712,
+      "loss": 0.644,
+      "step": 1180
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 0.0002576653696498054,
+      "loss": 0.6465,
+      "step": 1190
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 0.00025727626459143965,
+      "loss": 0.6685,
+      "step": 1200
+    },
+    {
+      "epoch": 0.77,
+      "eval_loss": 0.6553727984428406,
+      "eval_runtime": 303.8445,
+      "eval_samples_per_second": 6.582,
+      "eval_steps_per_second": 0.823,
+      "step": 1200
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 0.0002568871595330739,
+      "loss": 0.6359,
+      "step": 1210
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 0.00025649805447470813,
+      "loss": 0.639,
+      "step": 1220
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 0.0002561089494163424,
+      "loss": 0.6284,
+      "step": 1230
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 0.0002557198443579766,
+      "loss": 0.6625,
+      "step": 1240
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 0.00025533073929961085,
+      "loss": 0.6389,
+      "step": 1250
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 0.0002549416342412451,
+      "loss": 0.6784,
+      "step": 1260
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 0.00025455252918287933,
+      "loss": 0.6379,
+      "step": 1270
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 0.00025416342412451357,
+      "loss": 0.6553,
+      "step": 1280
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 0.0002537743190661478,
+      "loss": 0.6721,
+      "step": 1290
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 0.0002533852140077821,
+      "loss": 0.6442,
+      "step": 1300
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 0.00025299610894941634,
+      "loss": 0.642,
+      "step": 1310
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 0.0002526070038910506,
+      "loss": 0.6109,
+      "step": 1320
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 0.0002522178988326848,
+      "loss": 0.6829,
+      "step": 1330
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 0.00025182879377431906,
+      "loss": 0.634,
+      "step": 1340
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 0.0002514396887159533,
+      "loss": 0.6483,
+      "step": 1350
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 0.00025105058365758754,
+      "loss": 0.6571,
+      "step": 1360
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 0.0002506614785992218,
+      "loss": 0.6739,
+      "step": 1370
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 0.000250272373540856,
+      "loss": 0.6575,
+      "step": 1380
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 0.00024988326848249026,
+      "loss": 0.6446,
+      "step": 1390
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 0.0002494941634241245,
+      "loss": 0.6264,
+      "step": 1400
+    },
+    {
+      "epoch": 0.9,
+      "eval_loss": 0.6528891921043396,
+      "eval_runtime": 304.6435,
+      "eval_samples_per_second": 6.565,
+      "eval_steps_per_second": 0.821,
+      "step": 1400
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 0.00024910505836575874,
+      "loss": 0.6295,
+      "step": 1410
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 0.000248715953307393,
+      "loss": 0.6534,
+      "step": 1420
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 0.0002483268482490272,
+      "loss": 0.6448,
+      "step": 1430
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 0.00024793774319066146,
+      "loss": 0.6337,
+      "step": 1440
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 0.0002475486381322957,
+      "loss": 0.6344,
+      "step": 1450
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 0.00024715953307392994,
+      "loss": 0.6451,
+      "step": 1460
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 0.0002467704280155642,
+      "loss": 0.6562,
+      "step": 1470
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 0.0002463813229571984,
+      "loss": 0.6478,
+      "step": 1480
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 0.00024599221789883266,
+      "loss": 0.6532,
+      "step": 1490
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 0.0002456031128404669,
+      "loss": 0.6384,
+      "step": 1500
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 0.00024521400778210114,
+      "loss": 0.6422,
+      "step": 1510
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 0.0002448249027237354,
+      "loss": 0.6557,
+      "step": 1520
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 0.0002444357976653696,
+      "loss": 0.6405,
+      "step": 1530
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 0.00024404669260700388,
+      "loss": 0.6654,
+      "step": 1540
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 0.00024365758754863812,
+      "loss": 0.6676,
+      "step": 1550
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 0.00024326848249027236,
+      "loss": 0.6544,
+      "step": 1560
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 0.0002428793774319066,
+      "loss": 0.6426,
+      "step": 1570
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 0.00024249027237354084,
+      "loss": 0.6309,
+      "step": 1580
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 0.00024210116731517508,
+      "loss": 0.6543,
+      "step": 1590
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 0.00024171206225680932,
+      "loss": 0.6244,
+      "step": 1600
+    },
+    {
+      "epoch": 1.02,
+      "eval_loss": 0.652025043964386,
+      "eval_runtime": 302.2996,
+      "eval_samples_per_second": 6.616,
+      "eval_steps_per_second": 0.827,
+      "step": 1600
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 0.00024132295719844356,
+      "loss": 0.6328,
+      "step": 1610
+    },
+    {
+      "epoch": 1.04,
+      "learning_rate": 0.0002409338521400778,
+      "loss": 0.6316,
+      "step": 1620
+    },
+    {
+      "epoch": 1.04,
+      "learning_rate": 0.00024054474708171207,
+      "loss": 0.6558,
+      "step": 1630
+    },
+    {
+      "epoch": 1.05,
+      "learning_rate": 0.0002401556420233463,
+      "loss": 0.6512,
+      "step": 1640
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 0.00023976653696498054,
+      "loss": 0.6193,
+      "step": 1650
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 0.00023937743190661478,
+      "loss": 0.6447,
+      "step": 1660
+    },
+    {
+      "epoch": 1.07,
+      "learning_rate": 0.000238988326848249,
+      "loss": 0.6507,
+      "step": 1670
+    },
+    {
+      "epoch": 1.08,
+      "learning_rate": 0.00023859922178988324,
+      "loss": 0.6281,
+      "step": 1680
+    },
+    {
+      "epoch": 1.08,
+      "learning_rate": 0.00023821011673151748,
+      "loss": 0.6331,
+      "step": 1690
+    },
+    {
+      "epoch": 1.09,
+      "learning_rate": 0.00023782101167315172,
+      "loss": 0.6306,
+      "step": 1700
+    },
+    {
+      "epoch": 1.09,
+      "learning_rate": 0.00023743190661478595,
+      "loss": 0.6233,
+      "step": 1710
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 0.0002370428015564202,
+      "loss": 0.627,
+      "step": 1720
+    },
+    {
+      "epoch": 1.11,
+      "learning_rate": 0.00023665369649805443,
+      "loss": 0.6494,
+      "step": 1730
+    },
+    {
+      "epoch": 1.11,
+      "learning_rate": 0.0002362645914396887,
+      "loss": 0.6319,
+      "step": 1740
+    },
+    {
+      "epoch": 1.12,
+      "learning_rate": 0.00023587548638132294,
+      "loss": 0.6666,
+      "step": 1750
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 0.00023548638132295718,
+      "loss": 0.6097,
+      "step": 1760
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 0.00023509727626459142,
+      "loss": 0.6222,
+      "step": 1770
+    },
+    {
+      "epoch": 1.14,
+      "learning_rate": 0.00023470817120622566,
+      "loss": 0.6345,
+      "step": 1780
+    },
+    {
+      "epoch": 1.15,
+      "learning_rate": 0.0002343190661478599,
+      "loss": 0.6506,
+      "step": 1790
+    },
+    {
+      "epoch": 1.15,
+      "learning_rate": 0.00023392996108949414,
+      "loss": 0.6451,
+      "step": 1800
+    },
+    {
+      "epoch": 1.15,
+      "eval_loss": 0.6515942215919495,
+      "eval_runtime": 301.7791,
+      "eval_samples_per_second": 6.627,
+      "eval_steps_per_second": 0.828,
+      "step": 1800
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 0.00023354085603112838,
+      "loss": 0.664,
+      "step": 1810
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 0.00023315175097276262,
+      "loss": 0.6299,
+      "step": 1820
+    },
+    {
+      "epoch": 1.17,
+      "learning_rate": 0.00023276264591439688,
+      "loss": 0.6286,
+      "step": 1830
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 0.00023237354085603112,
+      "loss": 0.6216,
+      "step": 1840
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 0.00023198443579766536,
+      "loss": 0.652,
+      "step": 1850
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 0.0002315953307392996,
+      "loss": 0.6092,
+      "step": 1860
+    },
+    {
+      "epoch": 1.2,
+      "learning_rate": 0.00023120622568093384,
+      "loss": 0.6369,
+      "step": 1870
+    },
+    {
+      "epoch": 1.2,
+      "learning_rate": 0.00023081712062256808,
+      "loss": 0.6229,
+      "step": 1880
+    },
+    {
+      "epoch": 1.21,
+      "learning_rate": 0.00023042801556420232,
+      "loss": 0.6274,
+      "step": 1890
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 0.00023003891050583656,
+      "loss": 0.6472,
+      "step": 1900
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 0.00022964980544747077,
+      "loss": 0.6351,
+      "step": 1910
+    },
+    {
+      "epoch": 1.23,
+      "learning_rate": 0.000229260700389105,
+      "loss": 0.6154,
+      "step": 1920
+    },
+    {
+      "epoch": 1.24,
+      "learning_rate": 0.00022887159533073925,
+      "loss": 0.6322,
+      "step": 1930
+    },
+    {
+      "epoch": 1.24,
+      "learning_rate": 0.00022848249027237352,
+      "loss": 0.6323,
+      "step": 1940
+    },
+    {
+      "epoch": 1.25,
+      "learning_rate": 0.00022809338521400776,
+      "loss": 0.6222,
+      "step": 1950
+    },
+    {
+      "epoch": 1.25,
+      "learning_rate": 0.000227704280155642,
+      "loss": 0.6119,
+      "step": 1960
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 0.00022731517509727624,
+      "loss": 0.621,
+      "step": 1970
+    },
+    {
+      "epoch": 1.27,
+      "learning_rate": 0.00022692607003891048,
+      "loss": 0.6653,
+      "step": 1980
+    },
+    {
+      "epoch": 1.27,
+      "learning_rate": 0.00022653696498054472,
+      "loss": 0.6739,
+      "step": 1990
+    },
+    {
+      "epoch": 1.28,
+      "learning_rate": 0.00022614785992217896,
+      "loss": 0.6398,
+      "step": 2000
+    },
+    {
+      "epoch": 1.28,
+      "eval_loss": 0.6505751013755798,
+      "eval_runtime": 333.7471,
+      "eval_samples_per_second": 5.993,
+      "eval_steps_per_second": 0.749,
+      "step": 2000
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 0.0002257587548638132,
+      "loss": 0.6455,
+      "step": 2010
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 0.00022536964980544744,
+      "loss": 0.64,
+      "step": 2020
+    },
+    {
+      "epoch": 1.3,
+      "learning_rate": 0.0002249805447470817,
+      "loss": 0.6313,
+      "step": 2030
+    },
+    {
+      "epoch": 1.31,
+      "learning_rate": 0.00022459143968871594,
+      "loss": 0.64,
+      "step": 2040
+    },
+    {
+      "epoch": 1.31,
+      "learning_rate": 0.00022420233463035018,
+      "loss": 0.5932,
+      "step": 2050
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 0.00022381322957198442,
+      "loss": 0.6273,
+      "step": 2060
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 0.00022342412451361866,
+      "loss": 0.6347,
+      "step": 2070
+    },
+    {
+      "epoch": 1.33,
+      "learning_rate": 0.0002230350194552529,
+      "loss": 0.623,
+      "step": 2080
+    },
+    {
+      "epoch": 1.34,
+      "learning_rate": 0.00022264591439688714,
+      "loss": 0.6247,
+      "step": 2090
+    },
+    {
+      "epoch": 1.34,
+      "learning_rate": 0.00022225680933852138,
+      "loss": 0.6411,
+      "step": 2100
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 0.00022186770428015565,
+      "loss": 0.6342,
+      "step": 2110
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 0.0002214785992217899,
+      "loss": 0.6416,
+      "step": 2120
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 0.00022108949416342413,
+      "loss": 0.6068,
+      "step": 2130
+    },
+    {
+      "epoch": 1.37,
+      "learning_rate": 0.00022070038910505837,
+      "loss": 0.6255,
+      "step": 2140
+    },
+    {
+      "epoch": 1.38,
+      "learning_rate": 0.00022031128404669258,
+      "loss": 0.6522,
+      "step": 2150
+    },
+    {
+      "epoch": 1.38,
+      "learning_rate": 0.00021992217898832682,
+      "loss": 0.6289,
+      "step": 2160
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 0.00021953307392996106,
+      "loss": 0.651,
+      "step": 2170
+    },
+    {
+      "epoch": 1.4,
+      "learning_rate": 0.0002191439688715953,
+      "loss": 0.6257,
+      "step": 2180
+    },
+    {
+      "epoch": 1.4,
+      "learning_rate": 0.00021875486381322954,
+      "loss": 0.6186,
+      "step": 2190
+    },
+    {
+      "epoch": 1.41,
+      "learning_rate": 0.00021836575875486378,
+      "loss": 0.6141,
+      "step": 2200
+    },
+    {
+      "epoch": 1.41,
+      "eval_loss": 0.6505178213119507,
+      "eval_runtime": 305.7766,
+      "eval_samples_per_second": 6.541,
+      "eval_steps_per_second": 0.818,
+      "step": 2200
+    },
+    {
+      "epoch": 1.41,
+      "learning_rate": 0.00021797665369649802,
+      "loss": 0.6214,
+      "step": 2210
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 0.00021758754863813228,
+      "loss": 0.618,
+      "step": 2220
+    },
+    {
+      "epoch": 1.43,
+      "learning_rate": 0.00021719844357976652,
+      "loss": 0.612,
+      "step": 2230
+    },
+    {
+      "epoch": 1.43,
+      "learning_rate": 0.00021680933852140076,
+      "loss": 0.6406,
+      "step": 2240
+    },
+    {
+      "epoch": 1.44,
+      "learning_rate": 0.000216420233463035,
+      "loss": 0.6184,
+      "step": 2250
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 0.00021603112840466924,
+      "loss": 0.6354,
+      "step": 2260
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 0.00021564202334630348,
+      "loss": 0.6315,
+      "step": 2270
+    },
+    {
+      "epoch": 1.46,
+      "learning_rate": 0.00021525291828793772,
+      "loss": 0.6254,
+      "step": 2280
+    },
+    {
+      "epoch": 1.47,
+      "learning_rate": 0.00021486381322957196,
+      "loss": 0.6283,
+      "step": 2290
+    },
+    {
+      "epoch": 1.47,
+      "learning_rate": 0.0002144747081712062,
+      "loss": 0.6302,
+      "step": 2300
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 0.00021408560311284047,
+      "loss": 0.6234,
+      "step": 2310
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 0.0002136964980544747,
+      "loss": 0.6592,
+      "step": 2320
+    },
+    {
+      "epoch": 1.49,
+      "learning_rate": 0.00021330739299610895,
+      "loss": 0.6247,
+      "step": 2330
+    },
+    {
+      "epoch": 1.5,
+      "learning_rate": 0.00021291828793774318,
+      "loss": 0.6383,
+      "step": 2340
+    },
+    {
+      "epoch": 1.5,
+      "learning_rate": 0.00021252918287937742,
+      "loss": 0.6451,
+      "step": 2350
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 0.00021214007782101166,
+      "loss": 0.6311,
+      "step": 2360
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 0.0002117509727626459,
+      "loss": 0.6441,
+      "step": 2370
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 0.00021136186770428014,
+      "loss": 0.6405,
+      "step": 2380
+    },
+    {
+      "epoch": 1.53,
+      "learning_rate": 0.00021097276264591438,
+      "loss": 0.6447,
+      "step": 2390
+    },
+    {
+      "epoch": 1.54,
+      "learning_rate": 0.0002105836575875486,
+      "loss": 0.6311,
+      "step": 2400
+    },
+    {
+      "epoch": 1.54,
+      "eval_loss": 0.6494212746620178,
+      "eval_runtime": 305.7108,
+      "eval_samples_per_second": 6.542,
+      "eval_steps_per_second": 0.818,
+      "step": 2400
+    },
+    {
+      "epoch": 1.54,
+      "learning_rate": 0.00021019455252918283,
+      "loss": 0.6296,
+      "step": 2410
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 0.0002098054474708171,
+      "loss": 0.6444,
+      "step": 2420
+    },
+    {
+      "epoch": 1.56,
+      "learning_rate": 0.00020941634241245134,
+      "loss": 0.6366,
+      "step": 2430
+    },
+    {
+      "epoch": 1.56,
+      "learning_rate": 0.00020902723735408558,
+      "loss": 0.6433,
+      "step": 2440
+    },
+    {
+      "epoch": 1.57,
+      "learning_rate": 0.00020863813229571982,
+      "loss": 0.6302,
+      "step": 2450
+    },
+    {
+      "epoch": 1.57,
+      "learning_rate": 0.00020824902723735406,
+      "loss": 0.6134,
+      "step": 2460
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 0.0002078599221789883,
+      "loss": 0.617,
+      "step": 2470
+    },
+    {
+      "epoch": 1.59,
+      "learning_rate": 0.00020747081712062254,
+      "loss": 0.6574,
+      "step": 2480
+    },
+    {
+      "epoch": 1.59,
+      "learning_rate": 0.00020708171206225678,
+      "loss": 0.6267,
+      "step": 2490
+    },
+    {
+      "epoch": 1.6,
+      "learning_rate": 0.00020669260700389102,
+      "loss": 0.6151,
+      "step": 2500
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 0.00020630350194552528,
+      "loss": 0.66,
+      "step": 2510
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 0.00020591439688715952,
+      "loss": 0.6372,
+      "step": 2520
+    },
+    {
+      "epoch": 1.62,
+      "learning_rate": 0.00020552529182879376,
+      "loss": 0.6184,
+      "step": 2530
+    },
+    {
+      "epoch": 1.63,
+      "learning_rate": 0.000205136186770428,
+      "loss": 0.6115,
+      "step": 2540
+    },
+    {
+      "epoch": 1.63,
+      "learning_rate": 0.00020474708171206224,
+      "loss": 0.6616,
+      "step": 2550
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 0.00020435797665369648,
+      "loss": 0.6367,
+      "step": 2560
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 0.00020396887159533072,
+      "loss": 0.6464,
+      "step": 2570
+    },
+    {
+      "epoch": 1.65,
+      "learning_rate": 0.00020357976653696496,
+      "loss": 0.6304,
+      "step": 2580
+    },
+    {
+      "epoch": 1.66,
+      "learning_rate": 0.0002031906614785992,
+      "loss": 0.6015,
+      "step": 2590
+    },
+    {
+      "epoch": 1.66,
+      "learning_rate": 0.00020280155642023347,
+      "loss": 0.6353,
+      "step": 2600
+    },
+    {
+      "epoch": 1.66,
+      "eval_loss": 0.6482927203178406,
+      "eval_runtime": 309.3391,
+      "eval_samples_per_second": 6.465,
+      "eval_steps_per_second": 0.808,
+      "step": 2600
+    },
+    {
+      "epoch": 1.67,
+      "learning_rate": 0.0002024124513618677,
+      "loss": 0.6311,
+      "step": 2610
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 0.00020202334630350195,
+      "loss": 0.6221,
+      "step": 2620
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 0.0002016342412451362,
+      "loss": 0.6439,
+      "step": 2630
+    },
+    {
+      "epoch": 1.69,
+      "learning_rate": 0.0002012451361867704,
+      "loss": 0.6284,
+      "step": 2640
+    },
+    {
+      "epoch": 1.7,
+      "learning_rate": 0.00020085603112840464,
+      "loss": 0.6402,
+      "step": 2650
+    },
+    {
+      "epoch": 1.7,
+      "learning_rate": 0.00020046692607003888,
+      "loss": 0.6373,
+      "step": 2660
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 0.00020007782101167312,
+      "loss": 0.6205,
+      "step": 2670
+    },
+    {
+      "epoch": 1.72,
+      "learning_rate": 0.00019968871595330736,
+      "loss": 0.6343,
+      "step": 2680
+    },
+    {
+      "epoch": 1.72,
+      "learning_rate": 0.0001992996108949416,
+      "loss": 0.6272,
+      "step": 2690
+    },
+    {
+      "epoch": 1.73,
+      "learning_rate": 0.00019891050583657586,
+      "loss": 0.6262,
+      "step": 2700
+    },
+    {
+      "epoch": 1.73,
+      "learning_rate": 0.0001985214007782101,
+      "loss": 0.6295,
+      "step": 2710
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 0.00019813229571984434,
+      "loss": 0.6256,
+      "step": 2720
+    },
+    {
+      "epoch": 1.75,
+      "learning_rate": 0.00019774319066147858,
+      "loss": 0.6485,
+      "step": 2730
+    },
+    {
+      "epoch": 1.75,
+      "learning_rate": 0.00019735408560311282,
+      "loss": 0.6518,
+      "step": 2740
+    },
+    {
+      "epoch": 1.76,
+      "learning_rate": 0.00019696498054474706,
+      "loss": 0.6627,
+      "step": 2750
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 0.0001965758754863813,
+      "loss": 0.594,
+      "step": 2760
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 0.00019618677042801554,
+      "loss": 0.6289,
+      "step": 2770
+    },
+    {
+      "epoch": 1.78,
+      "learning_rate": 0.00019579766536964978,
+      "loss": 0.6393,
+      "step": 2780
+    },
+    {
+      "epoch": 1.79,
+      "learning_rate": 0.00019540856031128405,
+      "loss": 0.6073,
+      "step": 2790
+    },
+    {
+      "epoch": 1.79,
+      "learning_rate": 0.0001950194552529183,
+      "loss": 0.6331,
+      "step": 2800
+    },
+    {
+      "epoch": 1.79,
+      "eval_loss": 0.647841215133667,
+      "eval_runtime": 307.5546,
+      "eval_samples_per_second": 6.503,
+      "eval_steps_per_second": 0.813,
+      "step": 2800
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 0.00019463035019455253,
+      "loss": 0.6353,
+      "step": 2810
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 0.00019424124513618677,
+      "loss": 0.6444,
+      "step": 2820
+    },
+    {
+      "epoch": 1.81,
+      "learning_rate": 0.000193852140077821,
+      "loss": 0.6227,
+      "step": 2830
+    },
+    {
+      "epoch": 1.82,
+      "learning_rate": 0.00019346303501945525,
+      "loss": 0.6505,
+      "step": 2840
+    },
+    {
+      "epoch": 1.82,
+      "learning_rate": 0.00019307392996108949,
+      "loss": 0.6545,
+      "step": 2850
+    },
+    {
+      "epoch": 1.83,
+      "learning_rate": 0.00019268482490272372,
+      "loss": 0.6354,
+      "step": 2860
+    },
+    {
+      "epoch": 1.84,
+      "learning_rate": 0.00019229571984435796,
+      "loss": 0.5997,
+      "step": 2870
+    },
+    {
+      "epoch": 1.84,
+      "learning_rate": 0.00019190661478599218,
+      "loss": 0.6465,
+      "step": 2880
+    },
+    {
+      "epoch": 1.85,
+      "learning_rate": 0.00019151750972762642,
+      "loss": 0.6198,
+      "step": 2890
+    },
+    {
+      "epoch": 1.86,
+      "learning_rate": 0.00019112840466926068,
+      "loss": 0.6407,
+      "step": 2900
+    },
+    {
+      "epoch": 1.86,
+      "learning_rate": 0.00019073929961089492,
+      "loss": 0.6413,
+      "step": 2910
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 0.00019035019455252916,
+      "loss": 0.6468,
+      "step": 2920
+    },
+    {
+      "epoch": 1.88,
+      "learning_rate": 0.0001899610894941634,
+      "loss": 0.6133,
+      "step": 2930
+    },
+    {
+      "epoch": 1.88,
+      "learning_rate": 0.00018957198443579764,
+      "loss": 0.6055,
+      "step": 2940
+    },
+    {
+      "epoch": 1.89,
+      "learning_rate": 0.00018918287937743188,
+      "loss": 0.6155,
+      "step": 2950
+    },
+    {
+      "epoch": 1.89,
+      "learning_rate": 0.00018879377431906612,
+      "loss": 0.6253,
+      "step": 2960
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 0.00018840466926070036,
+      "loss": 0.631,
+      "step": 2970
+    },
+    {
+      "epoch": 1.91,
+      "learning_rate": 0.0001880155642023346,
+      "loss": 0.6332,
+      "step": 2980
+    },
+    {
+      "epoch": 1.91,
+      "learning_rate": 0.00018762645914396887,
+      "loss": 0.6229,
+      "step": 2990
+    },
+    {
+      "epoch": 1.92,
+      "learning_rate": 0.0001872373540856031,
+      "loss": 0.6291,
+      "step": 3000
+    },
+    {
+      "epoch": 1.92,
+      "eval_loss": 0.6470396518707275,
+      "eval_runtime": 303.5236,
+      "eval_samples_per_second": 6.589,
+      "eval_steps_per_second": 0.824,
+      "step": 3000
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 7810,
+  "num_train_epochs": 5,
+  "save_steps": 200,
+  "total_flos": 9.12950959125037e+17,
+  "trial_name": null,
+  "trial_params": null
+}

with_input/1a_52k/best_model/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:02d9b5d6dd2c93436238cd7f358b8b49214de02a02f7875becf0d9836de18aa7
+size 4027

with_input/1b_52k/best_model/adapter_config.json

@@ -0,0 +1,21 @@
+{
+  "base_model_name_or_path": "huggyllama/llama-7b",
+  "bias": "none",
+  "enable_lora": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "lora_alpha": 16,
+  "lora_dropout": 0.05,
+  "merge_weights": false,
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 16,
+  "target_modules": [
+    "q_proj",
+    "k_proj",
+    "v_proj",
+    "o_proj"
+  ],
+  "task_type": "CAUSAL_LM"
+}

with_input/1b_52k/best_model/adapter_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1f205dca0a65bf622caa6d439f4ff02151cac0a5c4ba8a1d3f4d6aad08d17898
+size 67201357

with_input/1b_52k/best_model/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1b5e6e20b1087f6644e6895c74078975f305a563890b6467d0cc6bcf54f2475a
+size 134433093

with_input/1b_52k/best_model/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d56b42bf9cdca4b0cc0311bae98becd374f3c91f842129455b13c79b24023f56
+size 14575

with_input/1b_52k/best_model/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:abfed23e6d6d347e99d70b8c08d200a84418499ce9da89b5fe35ff4e4e832571
+size 627

with_input/1b_52k/best_model/trainer_state.json

@@ -0,0 +1,2963 @@
+{
+  "best_metric": 0.4508701264858246,
+  "best_model_checkpoint": "checkpoints/instrucode/with_input/1b_52k/checkpoint-4600",
+  "epoch": 2.958199356913183,
+  "eval_steps": 200,
+  "global_step": 4600,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.6999999999999996e-05,
+      "loss": 1.2405,
+      "step": 10
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.6999999999999996e-05,
+      "loss": 1.2308,
+      "step": 20
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 8.699999999999999e-05,
+      "loss": 1.0636,
+      "step": 30
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 0.000117,
+      "loss": 0.8083,
+      "step": 40
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 0.000147,
+      "loss": 0.7358,
+      "step": 50
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 0.00017699999999999997,
+      "loss": 0.6593,
+      "step": 60
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 0.00020699999999999996,
+      "loss": 0.557,
+      "step": 70
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 0.000237,
+      "loss": 0.5436,
+      "step": 80
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 0.000267,
+      "loss": 0.5564,
+      "step": 90
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 0.00029699999999999996,
+      "loss": 0.5154,
+      "step": 100
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 0.00029964820846905535,
+      "loss": 0.5257,
+      "step": 110
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 0.000299257328990228,
+      "loss": 0.5174,
+      "step": 120
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 0.00029886644951140066,
+      "loss": 0.5303,
+      "step": 130
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 0.00029847557003257326,
+      "loss": 0.5081,
+      "step": 140
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 0.0002980846905537459,
+      "loss": 0.5031,
+      "step": 150
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 0.00029769381107491857,
+      "loss": 0.4832,
+      "step": 160
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 0.00029730293159609117,
+      "loss": 0.4949,
+      "step": 170
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 0.0002969120521172638,
+      "loss": 0.5122,
+      "step": 180
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 0.0002965211726384364,
+      "loss": 0.5149,
+      "step": 190
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 0.0002961302931596091,
+      "loss": 0.4987,
+      "step": 200
+    },
+    {
+      "epoch": 0.13,
+      "eval_loss": 0.49822479486465454,
+      "eval_runtime": 389.6569,
+      "eval_samples_per_second": 5.133,
+      "eval_steps_per_second": 0.642,
+      "step": 200
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 0.00029573941368078174,
+      "loss": 0.5059,
+      "step": 210
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 0.0002953485342019544,
+      "loss": 0.5117,
+      "step": 220
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 0.00029495765472312705,
+      "loss": 0.4844,
+      "step": 230
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 0.00029456677524429965,
+      "loss": 0.4967,
+      "step": 240
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 0.0002941758957654723,
+      "loss": 0.4805,
+      "step": 250
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 0.0002937850162866449,
+      "loss": 0.4713,
+      "step": 260
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 0.00029339413680781756,
+      "loss": 0.4857,
+      "step": 270
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 0.0002930032573289902,
+      "loss": 0.4935,
+      "step": 280
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 0.0002926123778501628,
+      "loss": 0.4827,
+      "step": 290
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 0.00029222149837133547,
+      "loss": 0.4954,
+      "step": 300
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 0.0002918306188925081,
+      "loss": 0.4902,
+      "step": 310
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 0.0002914397394136808,
+      "loss": 0.4707,
+      "step": 320
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 0.0002910488599348534,
+      "loss": 0.4909,
+      "step": 330
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 0.00029065798045602604,
+      "loss": 0.4763,
+      "step": 340
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 0.00029026710097719864,
+      "loss": 0.4896,
+      "step": 350
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 0.0002898762214983713,
+      "loss": 0.4778,
+      "step": 360
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 0.00028948534201954395,
+      "loss": 0.483,
+      "step": 370
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 0.0002890944625407166,
+      "loss": 0.4811,
+      "step": 380
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 0.0002887035830618892,
+      "loss": 0.4908,
+      "step": 390
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 0.00028831270358306186,
+      "loss": 0.478,
+      "step": 400
+    },
+    {
+      "epoch": 0.26,
+      "eval_loss": 0.48327910900115967,
+      "eval_runtime": 403.39,
+      "eval_samples_per_second": 4.958,
+      "eval_steps_per_second": 0.62,
+      "step": 400
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 0.0002879218241042345,
+      "loss": 0.4708,
+      "step": 410
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 0.0002875309446254071,
+      "loss": 0.4814,
+      "step": 420
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 0.0002871400651465798,
+      "loss": 0.4852,
+      "step": 430
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 0.00028674918566775243,
+      "loss": 0.4682,
+      "step": 440
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 0.00028635830618892503,
+      "loss": 0.4594,
+      "step": 450
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 0.0002859674267100977,
+      "loss": 0.4759,
+      "step": 460
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 0.00028557654723127034,
+      "loss": 0.4779,
+      "step": 470
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 0.000285185667752443,
+      "loss": 0.4739,
+      "step": 480
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 0.00028479478827361565,
+      "loss": 0.4698,
+      "step": 490
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 0.00028440390879478825,
+      "loss": 0.4819,
+      "step": 500
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 0.0002840130293159609,
+      "loss": 0.4768,
+      "step": 510
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 0.0002836221498371335,
+      "loss": 0.4705,
+      "step": 520
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 0.00028323127035830616,
+      "loss": 0.4755,
+      "step": 530
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 0.0002828403908794788,
+      "loss": 0.4686,
+      "step": 540
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 0.0002824495114006514,
+      "loss": 0.4745,
+      "step": 550
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 0.0002820586319218241,
+      "loss": 0.4889,
+      "step": 560
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 0.00028166775244299673,
+      "loss": 0.4696,
+      "step": 570
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 0.0002812768729641694,
+      "loss": 0.4896,
+      "step": 580
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 0.000280885993485342,
+      "loss": 0.4772,
+      "step": 590
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 0.00028049511400651464,
+      "loss": 0.4824,
+      "step": 600
+    },
+    {
+      "epoch": 0.39,
+      "eval_loss": 0.47541937232017517,
+      "eval_runtime": 392.9458,
+      "eval_samples_per_second": 5.09,
+      "eval_steps_per_second": 0.636,
+      "step": 600
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 0.00028010423452768724,
+      "loss": 0.481,
+      "step": 610
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 0.0002797133550488599,
+      "loss": 0.4695,
+      "step": 620
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 0.00027932247557003255,
+      "loss": 0.4725,
+      "step": 630
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 0.0002789315960912052,
+      "loss": 0.4688,
+      "step": 640
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 0.0002785407166123778,
+      "loss": 0.4782,
+      "step": 650
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 0.00027814983713355046,
+      "loss": 0.4775,
+      "step": 660
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 0.0002777589576547231,
+      "loss": 0.4702,
+      "step": 670
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 0.0002773680781758957,
+      "loss": 0.4606,
+      "step": 680
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 0.0002769771986970684,
+      "loss": 0.4625,
+      "step": 690
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 0.000276586319218241,
+      "loss": 0.4768,
+      "step": 700
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 0.00027619543973941363,
+      "loss": 0.4695,
+      "step": 710
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 0.0002758045602605863,
+      "loss": 0.472,
+      "step": 720
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 0.00027541368078175894,
+      "loss": 0.4786,
+      "step": 730
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 0.0002750228013029316,
+      "loss": 0.4832,
+      "step": 740
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 0.0002746319218241042,
+      "loss": 0.4788,
+      "step": 750
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 0.00027424104234527685,
+      "loss": 0.465,
+      "step": 760
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 0.0002738501628664495,
+      "loss": 0.4868,
+      "step": 770
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 0.0002734592833876221,
+      "loss": 0.4659,
+      "step": 780
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 0.00027306840390879477,
+      "loss": 0.4667,
+      "step": 790
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 0.00027267752442996737,
+      "loss": 0.4643,
+      "step": 800
+    },
+    {
+      "epoch": 0.51,
+      "eval_loss": 0.4702647626399994,
+      "eval_runtime": 395.8141,
+      "eval_samples_per_second": 5.053,
+      "eval_steps_per_second": 0.632,
+      "step": 800
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 0.00027228664495114,
+      "loss": 0.4593,
+      "step": 810
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 0.0002718957654723127,
+      "loss": 0.4602,
+      "step": 820
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 0.00027150488599348533,
+      "loss": 0.4663,
+      "step": 830
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 0.000271114006514658,
+      "loss": 0.4812,
+      "step": 840
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 0.0002707231270358306,
+      "loss": 0.4682,
+      "step": 850
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 0.00027033224755700324,
+      "loss": 0.4506,
+      "step": 860
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 0.00026994136807817584,
+      "loss": 0.4786,
+      "step": 870
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 0.0002695504885993485,
+      "loss": 0.4516,
+      "step": 880
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 0.00026915960912052116,
+      "loss": 0.4591,
+      "step": 890
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 0.0002687687296416938,
+      "loss": 0.4581,
+      "step": 900
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 0.0002683778501628664,
+      "loss": 0.4509,
+      "step": 910
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 0.00026798697068403907,
+      "loss": 0.475,
+      "step": 920
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 0.0002675960912052117,
+      "loss": 0.4828,
+      "step": 930
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 0.0002672052117263843,
+      "loss": 0.4663,
+      "step": 940
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 0.000266814332247557,
+      "loss": 0.4767,
+      "step": 950
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 0.0002664234527687296,
+      "loss": 0.4633,
+      "step": 960
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 0.00026603257328990223,
+      "loss": 0.4877,
+      "step": 970
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 0.0002656416938110749,
+      "loss": 0.4529,
+      "step": 980
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 0.00026525081433224755,
+      "loss": 0.4742,
+      "step": 990
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 0.0002648599348534202,
+      "loss": 0.468,
+      "step": 1000
+    },
+    {
+      "epoch": 0.64,
+      "eval_loss": 0.46662116050720215,
+      "eval_runtime": 395.1813,
+      "eval_samples_per_second": 5.061,
+      "eval_steps_per_second": 0.633,
+      "step": 1000
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 0.0002644690553745928,
+      "loss": 0.4739,
+      "step": 1010
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 0.00026407817589576546,
+      "loss": 0.4653,
+      "step": 1020
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 0.0002636872964169381,
+      "loss": 0.448,
+      "step": 1030
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 0.0002632964169381107,
+      "loss": 0.4752,
+      "step": 1040
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 0.00026290553745928337,
+      "loss": 0.4625,
+      "step": 1050
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 0.00026251465798045597,
+      "loss": 0.4717,
+      "step": 1060
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 0.0002621237785016286,
+      "loss": 0.4715,
+      "step": 1070
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 0.0002617328990228013,
+      "loss": 0.4846,
+      "step": 1080
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 0.00026134201954397393,
+      "loss": 0.4464,
+      "step": 1090
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 0.0002609511400651466,
+      "loss": 0.465,
+      "step": 1100
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 0.0002605602605863192,
+      "loss": 0.4637,
+      "step": 1110
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 0.00026016938110749185,
+      "loss": 0.4535,
+      "step": 1120
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 0.00025977850162866445,
+      "loss": 0.4752,
+      "step": 1130
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 0.0002593876221498371,
+      "loss": 0.4563,
+      "step": 1140
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 0.00025899674267100976,
+      "loss": 0.4621,
+      "step": 1150
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 0.00025860586319218236,
+      "loss": 0.4634,
+      "step": 1160
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 0.000258214983713355,
+      "loss": 0.4593,
+      "step": 1170
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 0.00025782410423452767,
+      "loss": 0.464,
+      "step": 1180
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 0.0002574332247557003,
+      "loss": 0.4588,
+      "step": 1190
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 0.0002570423452768729,
+      "loss": 0.4605,
+      "step": 1200
+    },
+    {
+      "epoch": 0.77,
+      "eval_loss": 0.46422210335731506,
+      "eval_runtime": 395.5692,
+      "eval_samples_per_second": 5.056,
+      "eval_steps_per_second": 0.632,
+      "step": 1200
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 0.0002566514657980456,
+      "loss": 0.4725,
+      "step": 1210
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 0.0002562605863192182,
+      "loss": 0.4552,
+      "step": 1220
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 0.00025586970684039084,
+      "loss": 0.4766,
+      "step": 1230
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 0.0002554788273615635,
+      "loss": 0.4615,
+      "step": 1240
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 0.00025508794788273615,
+      "loss": 0.447,
+      "step": 1250
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 0.0002546970684039088,
+      "loss": 0.4557,
+      "step": 1260
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 0.0002543061889250814,
+      "loss": 0.4758,
+      "step": 1270
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 0.00025391530944625406,
+      "loss": 0.4569,
+      "step": 1280
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 0.0002535244299674267,
+      "loss": 0.4772,
+      "step": 1290
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 0.0002531335504885993,
+      "loss": 0.4646,
+      "step": 1300
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 0.00025274267100977197,
+      "loss": 0.4583,
+      "step": 1310
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 0.00025235179153094457,
+      "loss": 0.4427,
+      "step": 1320
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 0.00025196091205211723,
+      "loss": 0.454,
+      "step": 1330
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 0.0002515700325732899,
+      "loss": 0.4564,
+      "step": 1340
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 0.00025117915309446254,
+      "loss": 0.4752,
+      "step": 1350
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 0.0002507882736156352,
+      "loss": 0.454,
+      "step": 1360
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 0.0002503973941368078,
+      "loss": 0.4518,
+      "step": 1370
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 0.00025000651465798045,
+      "loss": 0.4723,
+      "step": 1380
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 0.00024961563517915305,
+      "loss": 0.448,
+      "step": 1390
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 0.0002492247557003257,
+      "loss": 0.47,
+      "step": 1400
+    },
+    {
+      "epoch": 0.9,
+      "eval_loss": 0.46156543493270874,
+      "eval_runtime": 401.374,
+      "eval_samples_per_second": 4.983,
+      "eval_steps_per_second": 0.623,
+      "step": 1400
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 0.00024883387622149836,
+      "loss": 0.4864,
+      "step": 1410
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 0.00024844299674267096,
+      "loss": 0.4461,
+      "step": 1420
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 0.0002480521172638436,
+      "loss": 0.4608,
+      "step": 1430
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 0.00024766123778501627,
+      "loss": 0.4562,
+      "step": 1440
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 0.00024727035830618893,
+      "loss": 0.4753,
+      "step": 1450
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 0.00024687947882736153,
+      "loss": 0.4485,
+      "step": 1460
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 0.0002464885993485342,
+      "loss": 0.4752,
+      "step": 1470
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 0.0002460977198697068,
+      "loss": 0.457,
+      "step": 1480
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 0.00024570684039087944,
+      "loss": 0.4668,
+      "step": 1490
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 0.0002453159609120521,
+      "loss": 0.4655,
+      "step": 1500
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 0.00024492508143322475,
+      "loss": 0.4448,
+      "step": 1510
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 0.00024453420195439735,
+      "loss": 0.4525,
+      "step": 1520
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 0.00024414332247557,
+      "loss": 0.4414,
+      "step": 1530
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 0.00024375244299674263,
+      "loss": 0.4635,
+      "step": 1540
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 0.0002433615635179153,
+      "loss": 0.4837,
+      "step": 1550
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 0.00024297068403908795,
+      "loss": 0.4482,
+      "step": 1560
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 0.00024257980456026055,
+      "loss": 0.443,
+      "step": 1570
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 0.0002421889250814332,
+      "loss": 0.4539,
+      "step": 1580
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 0.00024179804560260583,
+      "loss": 0.4554,
+      "step": 1590
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 0.00024140716612377849,
+      "loss": 0.4459,
+      "step": 1600
+    },
+    {
+      "epoch": 1.03,
+      "eval_loss": 0.45973867177963257,
+      "eval_runtime": 402.8623,
+      "eval_samples_per_second": 4.964,
+      "eval_steps_per_second": 0.621,
+      "step": 1600
+    },
+    {
+      "epoch": 1.04,
+      "learning_rate": 0.00024101628664495114,
+      "loss": 0.4464,
+      "step": 1610
+    },
+    {
+      "epoch": 1.04,
+      "learning_rate": 0.00024062540716612377,
+      "loss": 0.4333,
+      "step": 1620
+    },
+    {
+      "epoch": 1.05,
+      "learning_rate": 0.0002402345276872964,
+      "loss": 0.4484,
+      "step": 1630
+    },
+    {
+      "epoch": 1.05,
+      "learning_rate": 0.00023984364820846902,
+      "loss": 0.4449,
+      "step": 1640
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 0.00023945276872964168,
+      "loss": 0.4403,
+      "step": 1650
+    },
+    {
+      "epoch": 1.07,
+      "learning_rate": 0.0002390618892508143,
+      "loss": 0.4382,
+      "step": 1660
+    },
+    {
+      "epoch": 1.07,
+      "learning_rate": 0.00023867100977198696,
+      "loss": 0.4467,
+      "step": 1670
+    },
+    {
+      "epoch": 1.08,
+      "learning_rate": 0.00023828013029315956,
+      "loss": 0.4335,
+      "step": 1680
+    },
+    {
+      "epoch": 1.09,
+      "learning_rate": 0.00023788925081433222,
+      "loss": 0.4477,
+      "step": 1690
+    },
+    {
+      "epoch": 1.09,
+      "learning_rate": 0.00023749837133550487,
+      "loss": 0.4468,
+      "step": 1700
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 0.0002371074918566775,
+      "loss": 0.4219,
+      "step": 1710
+    },
+    {
+      "epoch": 1.11,
+      "learning_rate": 0.00023671661237785016,
+      "loss": 0.4577,
+      "step": 1720
+    },
+    {
+      "epoch": 1.11,
+      "learning_rate": 0.00023632573289902276,
+      "loss": 0.4597,
+      "step": 1730
+    },
+    {
+      "epoch": 1.12,
+      "learning_rate": 0.00023593485342019541,
+      "loss": 0.4396,
+      "step": 1740
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 0.00023554397394136807,
+      "loss": 0.4559,
+      "step": 1750
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 0.0002351530944625407,
+      "loss": 0.4504,
+      "step": 1760
+    },
+    {
+      "epoch": 1.14,
+      "learning_rate": 0.00023476221498371335,
+      "loss": 0.4387,
+      "step": 1770
+    },
+    {
+      "epoch": 1.14,
+      "learning_rate": 0.00023437133550488595,
+      "loss": 0.4391,
+      "step": 1780
+    },
+    {
+      "epoch": 1.15,
+      "learning_rate": 0.0002339804560260586,
+      "loss": 0.4502,
+      "step": 1790
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 0.00023358957654723124,
+      "loss": 0.4358,
+      "step": 1800
+    },
+    {
+      "epoch": 1.16,
+      "eval_loss": 0.4591727554798126,
+      "eval_runtime": 407.7185,
+      "eval_samples_per_second": 4.905,
+      "eval_steps_per_second": 0.613,
+      "step": 1800
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 0.0002331986970684039,
+      "loss": 0.4466,
+      "step": 1810
+    },
+    {
+      "epoch": 1.17,
+      "learning_rate": 0.00023280781758957655,
+      "loss": 0.4487,
+      "step": 1820
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 0.00023241693811074915,
+      "loss": 0.429,
+      "step": 1830
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 0.0002320260586319218,
+      "loss": 0.4414,
+      "step": 1840
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 0.00023163517915309443,
+      "loss": 0.4586,
+      "step": 1850
+    },
+    {
+      "epoch": 1.2,
+      "learning_rate": 0.0002312442996742671,
+      "loss": 0.4363,
+      "step": 1860
+    },
+    {
+      "epoch": 1.2,
+      "learning_rate": 0.00023085342019543974,
+      "loss": 0.44,
+      "step": 1870
+    },
+    {
+      "epoch": 1.21,
+      "learning_rate": 0.00023046254071661234,
+      "loss": 0.4623,
+      "step": 1880
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 0.000230071661237785,
+      "loss": 0.4352,
+      "step": 1890
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 0.00022968078175895763,
+      "loss": 0.434,
+      "step": 1900
+    },
+    {
+      "epoch": 1.23,
+      "learning_rate": 0.00022928990228013028,
+      "loss": 0.4411,
+      "step": 1910
+    },
+    {
+      "epoch": 1.23,
+      "learning_rate": 0.0002288990228013029,
+      "loss": 0.454,
+      "step": 1920
+    },
+    {
+      "epoch": 1.24,
+      "learning_rate": 0.00022850814332247554,
+      "loss": 0.4238,
+      "step": 1930
+    },
+    {
+      "epoch": 1.25,
+      "learning_rate": 0.00022811726384364817,
+      "loss": 0.4485,
+      "step": 1940
+    },
+    {
+      "epoch": 1.25,
+      "learning_rate": 0.00022772638436482082,
+      "loss": 0.4397,
+      "step": 1950
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 0.00022733550488599348,
+      "loss": 0.43,
+      "step": 1960
+    },
+    {
+      "epoch": 1.27,
+      "learning_rate": 0.0002269446254071661,
+      "loss": 0.4491,
+      "step": 1970
+    },
+    {
+      "epoch": 1.27,
+      "learning_rate": 0.00022655374592833873,
+      "loss": 0.4432,
+      "step": 1980
+    },
+    {
+      "epoch": 1.28,
+      "learning_rate": 0.00022616286644951136,
+      "loss": 0.4213,
+      "step": 1990
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 0.00022577198697068402,
+      "loss": 0.4368,
+      "step": 2000
+    },
+    {
+      "epoch": 1.29,
+      "eval_loss": 0.45750632882118225,
+      "eval_runtime": 399.3044,
+      "eval_samples_per_second": 5.009,
+      "eval_steps_per_second": 0.626,
+      "step": 2000
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 0.00022538110749185667,
+      "loss": 0.4461,
+      "step": 2010
+    },
+    {
+      "epoch": 1.3,
+      "learning_rate": 0.0002249902280130293,
+      "loss": 0.4253,
+      "step": 2020
+    },
+    {
+      "epoch": 1.31,
+      "learning_rate": 0.00022459934853420196,
+      "loss": 0.4429,
+      "step": 2030
+    },
+    {
+      "epoch": 1.31,
+      "learning_rate": 0.00022420846905537456,
+      "loss": 0.4436,
+      "step": 2040
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 0.0002238175895765472,
+      "loss": 0.4418,
+      "step": 2050
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 0.00022342671009771984,
+      "loss": 0.4156,
+      "step": 2060
+    },
+    {
+      "epoch": 1.33,
+      "learning_rate": 0.0002230358306188925,
+      "loss": 0.439,
+      "step": 2070
+    },
+    {
+      "epoch": 1.34,
+      "learning_rate": 0.00022264495114006515,
+      "loss": 0.451,
+      "step": 2080
+    },
+    {
+      "epoch": 1.34,
+      "learning_rate": 0.00022225407166123775,
+      "loss": 0.4468,
+      "step": 2090
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 0.0002218631921824104,
+      "loss": 0.4423,
+      "step": 2100
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 0.00022147231270358304,
+      "loss": 0.4267,
+      "step": 2110
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 0.0002210814332247557,
+      "loss": 0.4329,
+      "step": 2120
+    },
+    {
+      "epoch": 1.37,
+      "learning_rate": 0.00022069055374592835,
+      "loss": 0.4417,
+      "step": 2130
+    },
+    {
+      "epoch": 1.38,
+      "learning_rate": 0.00022029967426710095,
+      "loss": 0.4544,
+      "step": 2140
+    },
+    {
+      "epoch": 1.38,
+      "learning_rate": 0.0002199087947882736,
+      "loss": 0.4294,
+      "step": 2150
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 0.00021951791530944623,
+      "loss": 0.4484,
+      "step": 2160
+    },
+    {
+      "epoch": 1.4,
+      "learning_rate": 0.00021912703583061889,
+      "loss": 0.4521,
+      "step": 2170
+    },
+    {
+      "epoch": 1.4,
+      "learning_rate": 0.00021873615635179151,
+      "loss": 0.4362,
+      "step": 2180
+    },
+    {
+      "epoch": 1.41,
+      "learning_rate": 0.00021834527687296414,
+      "loss": 0.4342,
+      "step": 2190
+    },
+    {
+      "epoch": 1.41,
+      "learning_rate": 0.00021795439739413677,
+      "loss": 0.4394,
+      "step": 2200
+    },
+    {
+      "epoch": 1.41,
+      "eval_loss": 0.4563552439212799,
+      "eval_runtime": 401.35,
+      "eval_samples_per_second": 4.983,
+      "eval_steps_per_second": 0.623,
+      "step": 2200
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 0.00021756351791530943,
+      "loss": 0.444,
+      "step": 2210
+    },
+    {
+      "epoch": 1.43,
+      "learning_rate": 0.00021717263843648208,
+      "loss": 0.4434,
+      "step": 2220
+    },
+    {
+      "epoch": 1.43,
+      "learning_rate": 0.0002167817589576547,
+      "loss": 0.4451,
+      "step": 2230
+    },
+    {
+      "epoch": 1.44,
+      "learning_rate": 0.00021639087947882734,
+      "loss": 0.4347,
+      "step": 2240
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 0.00021599999999999996,
+      "loss": 0.4347,
+      "step": 2250
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 0.00021560912052117262,
+      "loss": 0.4532,
+      "step": 2260
+    },
+    {
+      "epoch": 1.46,
+      "learning_rate": 0.00021521824104234528,
+      "loss": 0.4578,
+      "step": 2270
+    },
+    {
+      "epoch": 1.47,
+      "learning_rate": 0.0002148273615635179,
+      "loss": 0.4479,
+      "step": 2280
+    },
+    {
+      "epoch": 1.47,
+      "learning_rate": 0.00021443648208469053,
+      "loss": 0.4606,
+      "step": 2290
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 0.00021404560260586316,
+      "loss": 0.431,
+      "step": 2300
+    },
+    {
+      "epoch": 1.49,
+      "learning_rate": 0.00021365472312703581,
+      "loss": 0.4584,
+      "step": 2310
+    },
+    {
+      "epoch": 1.49,
+      "learning_rate": 0.00021326384364820844,
+      "loss": 0.4459,
+      "step": 2320
+    },
+    {
+      "epoch": 1.5,
+      "learning_rate": 0.0002128729641693811,
+      "loss": 0.4469,
+      "step": 2330
+    },
+    {
+      "epoch": 1.5,
+      "learning_rate": 0.0002124820846905537,
+      "loss": 0.4335,
+      "step": 2340
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 0.00021209120521172635,
+      "loss": 0.4467,
+      "step": 2350
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 0.000211700325732899,
+      "loss": 0.4325,
+      "step": 2360
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 0.00021130944625407164,
+      "loss": 0.447,
+      "step": 2370
+    },
+    {
+      "epoch": 1.53,
+      "learning_rate": 0.0002109185667752443,
+      "loss": 0.4314,
+      "step": 2380
+    },
+    {
+      "epoch": 1.54,
+      "learning_rate": 0.00021052768729641695,
+      "loss": 0.4451,
+      "step": 2390
+    },
+    {
+      "epoch": 1.54,
+      "learning_rate": 0.00021013680781758955,
+      "loss": 0.4548,
+      "step": 2400
+    },
+    {
+      "epoch": 1.54,
+      "eval_loss": 0.4556201696395874,
+      "eval_runtime": 401.3952,
+      "eval_samples_per_second": 4.983,
+      "eval_steps_per_second": 0.623,
+      "step": 2400
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 0.0002097459283387622,
+      "loss": 0.4303,
+      "step": 2410
+    },
+    {
+      "epoch": 1.56,
+      "learning_rate": 0.00020935504885993483,
+      "loss": 0.4248,
+      "step": 2420
+    },
+    {
+      "epoch": 1.56,
+      "learning_rate": 0.0002089641693811075,
+      "loss": 0.4205,
+      "step": 2430
+    },
+    {
+      "epoch": 1.57,
+      "learning_rate": 0.00020857328990228012,
+      "loss": 0.4434,
+      "step": 2440
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 0.00020818241042345274,
+      "loss": 0.4544,
+      "step": 2450
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 0.00020779153094462537,
+      "loss": 0.4467,
+      "step": 2460
+    },
+    {
+      "epoch": 1.59,
+      "learning_rate": 0.00020740065146579803,
+      "loss": 0.4448,
+      "step": 2470
+    },
+    {
+      "epoch": 1.59,
+      "learning_rate": 0.00020700977198697068,
+      "loss": 0.4384,
+      "step": 2480
+    },
+    {
+      "epoch": 1.6,
+      "learning_rate": 0.0002066188925081433,
+      "loss": 0.4623,
+      "step": 2490
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 0.00020622801302931594,
+      "loss": 0.4413,
+      "step": 2500
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 0.00020583713355048857,
+      "loss": 0.4557,
+      "step": 2510
+    },
+    {
+      "epoch": 1.62,
+      "learning_rate": 0.00020544625407166122,
+      "loss": 0.4339,
+      "step": 2520
+    },
+    {
+      "epoch": 1.63,
+      "learning_rate": 0.00020505537459283388,
+      "loss": 0.4239,
+      "step": 2530
+    },
+    {
+      "epoch": 1.63,
+      "learning_rate": 0.0002046644951140065,
+      "loss": 0.4323,
+      "step": 2540
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 0.00020427361563517913,
+      "loss": 0.4479,
+      "step": 2550
+    },
+    {
+      "epoch": 1.65,
+      "learning_rate": 0.00020388273615635176,
+      "loss": 0.4448,
+      "step": 2560
+    },
+    {
+      "epoch": 1.65,
+      "learning_rate": 0.00020349185667752442,
+      "loss": 0.4348,
+      "step": 2570
+    },
+    {
+      "epoch": 1.66,
+      "learning_rate": 0.00020310097719869705,
+      "loss": 0.4344,
+      "step": 2580
+    },
+    {
+      "epoch": 1.67,
+      "learning_rate": 0.0002027100977198697,
+      "loss": 0.4401,
+      "step": 2590
+    },
+    {
+      "epoch": 1.67,
+      "learning_rate": 0.0002023192182410423,
+      "loss": 0.4366,
+      "step": 2600
+    },
+    {
+      "epoch": 1.67,
+      "eval_loss": 0.45435893535614014,
+      "eval_runtime": 403.3395,
+      "eval_samples_per_second": 4.959,
+      "eval_steps_per_second": 0.62,
+      "step": 2600
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 0.00020192833876221496,
+      "loss": 0.4367,
+      "step": 2610
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 0.0002015374592833876,
+      "loss": 0.4538,
+      "step": 2620
+    },
+    {
+      "epoch": 1.69,
+      "learning_rate": 0.00020114657980456024,
+      "loss": 0.4388,
+      "step": 2630
+    },
+    {
+      "epoch": 1.7,
+      "learning_rate": 0.0002007557003257329,
+      "loss": 0.4264,
+      "step": 2640
+    },
+    {
+      "epoch": 1.7,
+      "learning_rate": 0.0002003648208469055,
+      "loss": 0.4315,
+      "step": 2650
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 0.00019997394136807815,
+      "loss": 0.4304,
+      "step": 2660
+    },
+    {
+      "epoch": 1.72,
+      "learning_rate": 0.0001995830618892508,
+      "loss": 0.4332,
+      "step": 2670
+    },
+    {
+      "epoch": 1.72,
+      "learning_rate": 0.00019919218241042344,
+      "loss": 0.4431,
+      "step": 2680
+    },
+    {
+      "epoch": 1.73,
+      "learning_rate": 0.0001988013029315961,
+      "loss": 0.4386,
+      "step": 2690
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 0.0001984104234527687,
+      "loss": 0.4556,
+      "step": 2700
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 0.00019801954397394135,
+      "loss": 0.4463,
+      "step": 2710
+    },
+    {
+      "epoch": 1.75,
+      "learning_rate": 0.00019762866449511398,
+      "loss": 0.4496,
+      "step": 2720
+    },
+    {
+      "epoch": 1.76,
+      "learning_rate": 0.00019723778501628663,
+      "loss": 0.4313,
+      "step": 2730
+    },
+    {
+      "epoch": 1.76,
+      "learning_rate": 0.00019684690553745929,
+      "loss": 0.4378,
+      "step": 2740
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 0.00019645602605863191,
+      "loss": 0.4352,
+      "step": 2750
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 0.00019606514657980454,
+      "loss": 0.4385,
+      "step": 2760
+    },
+    {
+      "epoch": 1.78,
+      "learning_rate": 0.00019567426710097717,
+      "loss": 0.4312,
+      "step": 2770
+    },
+    {
+      "epoch": 1.79,
+      "learning_rate": 0.00019528338762214983,
+      "loss": 0.4339,
+      "step": 2780
+    },
+    {
+      "epoch": 1.79,
+      "learning_rate": 0.00019489250814332248,
+      "loss": 0.4609,
+      "step": 2790
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 0.0001945016286644951,
+      "loss": 0.4428,
+      "step": 2800
+    },
+    {
+      "epoch": 1.8,
+      "eval_loss": 0.45374733209609985,
+      "eval_runtime": 397.5954,
+      "eval_samples_per_second": 5.03,
+      "eval_steps_per_second": 0.629,
+      "step": 2800
+    },
+    {
+      "epoch": 1.81,
+      "learning_rate": 0.00019411074918566774,
+      "loss": 0.4311,
+      "step": 2810
+    },
+    {
+      "epoch": 1.81,
+      "learning_rate": 0.00019371986970684037,
+      "loss": 0.45,
+      "step": 2820
+    },
+    {
+      "epoch": 1.82,
+      "learning_rate": 0.00019332899022801302,
+      "loss": 0.4485,
+      "step": 2830
+    },
+    {
+      "epoch": 1.83,
+      "learning_rate": 0.00019293811074918565,
+      "loss": 0.4488,
+      "step": 2840
+    },
+    {
+      "epoch": 1.83,
+      "learning_rate": 0.0001925472312703583,
+      "loss": 0.4474,
+      "step": 2850
+    },
+    {
+      "epoch": 1.84,
+      "learning_rate": 0.0001921563517915309,
+      "loss": 0.4434,
+      "step": 2860
+    },
+    {
+      "epoch": 1.85,
+      "learning_rate": 0.00019176547231270356,
+      "loss": 0.4237,
+      "step": 2870
+    },
+    {
+      "epoch": 1.85,
+      "learning_rate": 0.00019137459283387622,
+      "loss": 0.4413,
+      "step": 2880
+    },
+    {
+      "epoch": 1.86,
+      "learning_rate": 0.00019098371335504884,
+      "loss": 0.444,
+      "step": 2890
+    },
+    {
+      "epoch": 1.86,
+      "learning_rate": 0.0001905928338762215,
+      "loss": 0.4351,
+      "step": 2900
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 0.0001902019543973941,
+      "loss": 0.4528,
+      "step": 2910
+    },
+    {
+      "epoch": 1.88,
+      "learning_rate": 0.00018981107491856675,
+      "loss": 0.4369,
+      "step": 2920
+    },
+    {
+      "epoch": 1.88,
+      "learning_rate": 0.0001894201954397394,
+      "loss": 0.4209,
+      "step": 2930
+    },
+    {
+      "epoch": 1.89,
+      "learning_rate": 0.00018902931596091204,
+      "loss": 0.4513,
+      "step": 2940
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 0.0001886384364820847,
+      "loss": 0.4353,
+      "step": 2950
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 0.0001882475570032573,
+      "loss": 0.438,
+      "step": 2960
+    },
+    {
+      "epoch": 1.91,
+      "learning_rate": 0.00018785667752442995,
+      "loss": 0.4522,
+      "step": 2970
+    },
+    {
+      "epoch": 1.92,
+      "learning_rate": 0.00018746579804560258,
+      "loss": 0.4303,
+      "step": 2980
+    },
+    {
+      "epoch": 1.92,
+      "learning_rate": 0.00018707491856677523,
+      "loss": 0.4698,
+      "step": 2990
+    },
+    {
+      "epoch": 1.93,
+      "learning_rate": 0.0001866840390879479,
+      "loss": 0.4386,
+      "step": 3000
+    },
+    {
+      "epoch": 1.93,
+      "eval_loss": 0.45254799723625183,
+      "eval_runtime": 392.9711,
+      "eval_samples_per_second": 5.089,
+      "eval_steps_per_second": 0.636,
+      "step": 3000
+    },
+    {
+      "epoch": 1.94,
+      "learning_rate": 0.0001862931596091205,
+      "loss": 0.4681,
+      "step": 3010
+    },
+    {
+      "epoch": 1.94,
+      "learning_rate": 0.00018590228013029314,
+      "loss": 0.4451,
+      "step": 3020
+    },
+    {
+      "epoch": 1.95,
+      "learning_rate": 0.00018551140065146577,
+      "loss": 0.4464,
+      "step": 3030
+    },
+    {
+      "epoch": 1.95,
+      "learning_rate": 0.00018512052117263843,
+      "loss": 0.4556,
+      "step": 3040
+    },
+    {
+      "epoch": 1.96,
+      "learning_rate": 0.00018472964169381108,
+      "loss": 0.4266,
+      "step": 3050
+    },
+    {
+      "epoch": 1.97,
+      "learning_rate": 0.00018433876221498368,
+      "loss": 0.4527,
+      "step": 3060
+    },
+    {
+      "epoch": 1.97,
+      "learning_rate": 0.00018394788273615634,
+      "loss": 0.4418,
+      "step": 3070
+    },
+    {
+      "epoch": 1.98,
+      "learning_rate": 0.00018355700325732897,
+      "loss": 0.4444,
+      "step": 3080
+    },
+    {
+      "epoch": 1.99,
+      "learning_rate": 0.00018316612377850162,
+      "loss": 0.4578,
+      "step": 3090
+    },
+    {
+      "epoch": 1.99,
+      "learning_rate": 0.00018277524429967425,
+      "loss": 0.4311,
+      "step": 3100
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 0.00018238436482084688,
+      "loss": 0.4411,
+      "step": 3110
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 0.0001819934853420195,
+      "loss": 0.4216,
+      "step": 3120
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 0.00018160260586319216,
+      "loss": 0.4057,
+      "step": 3130
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 0.00018121172638436482,
+      "loss": 0.4127,
+      "step": 3140
+    },
+    {
+      "epoch": 2.03,
+      "learning_rate": 0.00018082084690553745,
+      "loss": 0.4325,
+      "step": 3150
+    },
+    {
+      "epoch": 2.03,
+      "learning_rate": 0.0001804299674267101,
+      "loss": 0.4169,
+      "step": 3160
+    },
+    {
+      "epoch": 2.04,
+      "learning_rate": 0.0001800390879478827,
+      "loss": 0.4052,
+      "step": 3170
+    },
+    {
+      "epoch": 2.05,
+      "learning_rate": 0.00017964820846905536,
+      "loss": 0.4059,
+      "step": 3180
+    },
+    {
+      "epoch": 2.05,
+      "learning_rate": 0.000179257328990228,
+      "loss": 0.4234,
+      "step": 3190
+    },
+    {
+      "epoch": 2.06,
+      "learning_rate": 0.00017886644951140064,
+      "loss": 0.4145,
+      "step": 3200
+    },
+    {
+      "epoch": 2.06,
+      "eval_loss": 0.4534437358379364,
+      "eval_runtime": 397.6364,
+      "eval_samples_per_second": 5.03,
+      "eval_steps_per_second": 0.629,
+      "step": 3200
+    },
+    {
+      "epoch": 2.06,
+      "learning_rate": 0.0001784755700325733,
+      "loss": 0.4233,
+      "step": 3210
+    },
+    {
+      "epoch": 2.07,
+      "learning_rate": 0.0001780846905537459,
+      "loss": 0.3946,
+      "step": 3220
+    },
+    {
+      "epoch": 2.08,
+      "learning_rate": 0.00017769381107491855,
+      "loss": 0.4178,
+      "step": 3230
+    },
+    {
+      "epoch": 2.08,
+      "learning_rate": 0.00017730293159609118,
+      "loss": 0.4335,
+      "step": 3240
+    },
+    {
+      "epoch": 2.09,
+      "learning_rate": 0.00017691205211726384,
+      "loss": 0.4059,
+      "step": 3250
+    },
+    {
+      "epoch": 2.1,
+      "learning_rate": 0.0001765211726384365,
+      "loss": 0.4228,
+      "step": 3260
+    },
+    {
+      "epoch": 2.1,
+      "learning_rate": 0.0001761302931596091,
+      "loss": 0.4188,
+      "step": 3270
+    },
+    {
+      "epoch": 2.11,
+      "learning_rate": 0.00017573941368078175,
+      "loss": 0.4234,
+      "step": 3280
+    },
+    {
+      "epoch": 2.12,
+      "learning_rate": 0.00017534853420195438,
+      "loss": 0.411,
+      "step": 3290
+    },
+    {
+      "epoch": 2.12,
+      "learning_rate": 0.00017495765472312703,
+      "loss": 0.4268,
+      "step": 3300
+    },
+    {
+      "epoch": 2.13,
+      "learning_rate": 0.00017456677524429966,
+      "loss": 0.4105,
+      "step": 3310
+    },
+    {
+      "epoch": 2.14,
+      "learning_rate": 0.0001741758957654723,
+      "loss": 0.4298,
+      "step": 3320
+    },
+    {
+      "epoch": 2.14,
+      "learning_rate": 0.00017378501628664494,
+      "loss": 0.4178,
+      "step": 3330
+    },
+    {
+      "epoch": 2.15,
+      "learning_rate": 0.00017339413680781757,
+      "loss": 0.4222,
+      "step": 3340
+    },
+    {
+      "epoch": 2.15,
+      "learning_rate": 0.00017300325732899023,
+      "loss": 0.4199,
+      "step": 3350
+    },
+    {
+      "epoch": 2.16,
+      "learning_rate": 0.00017261237785016285,
+      "loss": 0.4258,
+      "step": 3360
+    },
+    {
+      "epoch": 2.17,
+      "learning_rate": 0.00017222149837133548,
+      "loss": 0.4154,
+      "step": 3370
+    },
+    {
+      "epoch": 2.17,
+      "learning_rate": 0.0001718306188925081,
+      "loss": 0.4105,
+      "step": 3380
+    },
+    {
+      "epoch": 2.18,
+      "learning_rate": 0.00017143973941368077,
+      "loss": 0.4259,
+      "step": 3390
+    },
+    {
+      "epoch": 2.19,
+      "learning_rate": 0.00017104885993485342,
+      "loss": 0.4048,
+      "step": 3400
+    },
+    {
+      "epoch": 2.19,
+      "eval_loss": 0.45390984416007996,
+      "eval_runtime": 397.2706,
+      "eval_samples_per_second": 5.034,
+      "eval_steps_per_second": 0.629,
+      "step": 3400
+    },
+    {
+      "epoch": 2.19,
+      "learning_rate": 0.00017065798045602605,
+      "loss": 0.4078,
+      "step": 3410
+    },
+    {
+      "epoch": 2.2,
+      "learning_rate": 0.00017026710097719868,
+      "loss": 0.413,
+      "step": 3420
+    },
+    {
+      "epoch": 2.21,
+      "learning_rate": 0.0001698762214983713,
+      "loss": 0.4175,
+      "step": 3430
+    },
+    {
+      "epoch": 2.21,
+      "learning_rate": 0.00016948534201954396,
+      "loss": 0.419,
+      "step": 3440
+    },
+    {
+      "epoch": 2.22,
+      "learning_rate": 0.0001690944625407166,
+      "loss": 0.4286,
+      "step": 3450
+    },
+    {
+      "epoch": 2.23,
+      "learning_rate": 0.00016870358306188924,
+      "loss": 0.4179,
+      "step": 3460
+    },
+    {
+      "epoch": 2.23,
+      "learning_rate": 0.00016831270358306187,
+      "loss": 0.4121,
+      "step": 3470
+    },
+    {
+      "epoch": 2.24,
+      "learning_rate": 0.0001679218241042345,
+      "loss": 0.4238,
+      "step": 3480
+    },
+    {
+      "epoch": 2.24,
+      "learning_rate": 0.00016753094462540716,
+      "loss": 0.4247,
+      "step": 3490
+    },
+    {
+      "epoch": 2.25,
+      "learning_rate": 0.00016714006514657978,
+      "loss": 0.4181,
+      "step": 3500
+    },
+    {
+      "epoch": 2.26,
+      "learning_rate": 0.00016674918566775244,
+      "loss": 0.4175,
+      "step": 3510
+    },
+    {
+      "epoch": 2.26,
+      "learning_rate": 0.0001663583061889251,
+      "loss": 0.4338,
+      "step": 3520
+    },
+    {
+      "epoch": 2.27,
+      "learning_rate": 0.0001659674267100977,
+      "loss": 0.4131,
+      "step": 3530
+    },
+    {
+      "epoch": 2.28,
+      "learning_rate": 0.00016557654723127035,
+      "loss": 0.4215,
+      "step": 3540
+    },
+    {
+      "epoch": 2.28,
+      "learning_rate": 0.00016518566775244298,
+      "loss": 0.4366,
+      "step": 3550
+    },
+    {
+      "epoch": 2.29,
+      "learning_rate": 0.00016479478827361563,
+      "loss": 0.424,
+      "step": 3560
+    },
+    {
+      "epoch": 2.3,
+      "learning_rate": 0.00016440390879478826,
+      "loss": 0.4147,
+      "step": 3570
+    },
+    {
+      "epoch": 2.3,
+      "learning_rate": 0.0001640130293159609,
+      "loss": 0.4223,
+      "step": 3580
+    },
+    {
+      "epoch": 2.31,
+      "learning_rate": 0.00016362214983713352,
+      "loss": 0.4132,
+      "step": 3590
+    },
+    {
+      "epoch": 2.32,
+      "learning_rate": 0.00016323127035830617,
+      "loss": 0.4019,
+      "step": 3600
+    },
+    {
+      "epoch": 2.32,
+      "eval_loss": 0.45374155044555664,
+      "eval_runtime": 403.7993,
+      "eval_samples_per_second": 4.953,
+      "eval_steps_per_second": 0.619,
+      "step": 3600
+    },
+    {
+      "epoch": 2.32,
+      "learning_rate": 0.00016284039087947883,
+      "loss": 0.4101,
+      "step": 3610
+    },
+    {
+      "epoch": 2.33,
+      "learning_rate": 0.00016244951140065146,
+      "loss": 0.4234,
+      "step": 3620
+    },
+    {
+      "epoch": 2.33,
+      "learning_rate": 0.00016205863192182408,
+      "loss": 0.4177,
+      "step": 3630
+    },
+    {
+      "epoch": 2.34,
+      "learning_rate": 0.0001616677524429967,
+      "loss": 0.433,
+      "step": 3640
+    },
+    {
+      "epoch": 2.35,
+      "learning_rate": 0.00016127687296416937,
+      "loss": 0.4249,
+      "step": 3650
+    },
+    {
+      "epoch": 2.35,
+      "learning_rate": 0.00016088599348534202,
+      "loss": 0.4233,
+      "step": 3660
+    },
+    {
+      "epoch": 2.36,
+      "learning_rate": 0.00016049511400651465,
+      "loss": 0.4229,
+      "step": 3670
+    },
+    {
+      "epoch": 2.37,
+      "learning_rate": 0.00016010423452768728,
+      "loss": 0.4221,
+      "step": 3680
+    },
+    {
+      "epoch": 2.37,
+      "learning_rate": 0.0001597133550488599,
+      "loss": 0.4166,
+      "step": 3690
+    },
+    {
+      "epoch": 2.38,
+      "learning_rate": 0.00015932247557003256,
+      "loss": 0.4183,
+      "step": 3700
+    },
+    {
+      "epoch": 2.39,
+      "learning_rate": 0.0001589315960912052,
+      "loss": 0.4279,
+      "step": 3710
+    },
+    {
+      "epoch": 2.39,
+      "learning_rate": 0.00015854071661237785,
+      "loss": 0.4246,
+      "step": 3720
+    },
+    {
+      "epoch": 2.4,
+      "learning_rate": 0.00015814983713355045,
+      "loss": 0.4225,
+      "step": 3730
+    },
+    {
+      "epoch": 2.41,
+      "learning_rate": 0.0001577589576547231,
+      "loss": 0.4199,
+      "step": 3740
+    },
+    {
+      "epoch": 2.41,
+      "learning_rate": 0.00015736807817589576,
+      "loss": 0.3944,
+      "step": 3750
+    },
+    {
+      "epoch": 2.42,
+      "learning_rate": 0.00015697719869706839,
+      "loss": 0.4287,
+      "step": 3760
+    },
+    {
+      "epoch": 2.42,
+      "learning_rate": 0.00015658631921824104,
+      "loss": 0.423,
+      "step": 3770
+    },
+    {
+      "epoch": 2.43,
+      "learning_rate": 0.00015619543973941364,
+      "loss": 0.4218,
+      "step": 3780
+    },
+    {
+      "epoch": 2.44,
+      "learning_rate": 0.0001558045602605863,
+      "loss": 0.4107,
+      "step": 3790
+    },
+    {
+      "epoch": 2.44,
+      "learning_rate": 0.00015541368078175895,
+      "loss": 0.4177,
+      "step": 3800
+    },
+    {
+      "epoch": 2.44,
+      "eval_loss": 0.4529287815093994,
+      "eval_runtime": 405.0457,
+      "eval_samples_per_second": 4.938,
+      "eval_steps_per_second": 0.617,
+      "step": 3800
+    },
+    {
+      "epoch": 2.45,
+      "learning_rate": 0.00015502280130293158,
+      "loss": 0.4184,
+      "step": 3810
+    },
+    {
+      "epoch": 2.46,
+      "learning_rate": 0.00015463192182410424,
+      "loss": 0.418,
+      "step": 3820
+    },
+    {
+      "epoch": 2.46,
+      "learning_rate": 0.00015424104234527684,
+      "loss": 0.417,
+      "step": 3830
+    },
+    {
+      "epoch": 2.47,
+      "learning_rate": 0.0001538501628664495,
+      "loss": 0.4262,
+      "step": 3840
+    },
+    {
+      "epoch": 2.48,
+      "learning_rate": 0.00015345928338762212,
+      "loss": 0.4191,
+      "step": 3850
+    },
+    {
+      "epoch": 2.48,
+      "learning_rate": 0.00015306840390879478,
+      "loss": 0.4301,
+      "step": 3860
+    },
+    {
+      "epoch": 2.49,
+      "learning_rate": 0.00015267752442996743,
+      "loss": 0.4129,
+      "step": 3870
+    },
+    {
+      "epoch": 2.5,
+      "learning_rate": 0.00015228664495114003,
+      "loss": 0.4272,
+      "step": 3880
+    },
+    {
+      "epoch": 2.5,
+      "learning_rate": 0.0001518957654723127,
+      "loss": 0.4196,
+      "step": 3890
+    },
+    {
+      "epoch": 2.51,
+      "learning_rate": 0.00015150488599348532,
+      "loss": 0.4083,
+      "step": 3900
+    },
+    {
+      "epoch": 2.51,
+      "learning_rate": 0.00015111400651465797,
+      "loss": 0.43,
+      "step": 3910
+    },
+    {
+      "epoch": 2.52,
+      "learning_rate": 0.00015072312703583063,
+      "loss": 0.4196,
+      "step": 3920
+    },
+    {
+      "epoch": 2.53,
+      "learning_rate": 0.00015033224755700325,
+      "loss": 0.4373,
+      "step": 3930
+    },
+    {
+      "epoch": 2.53,
+      "learning_rate": 0.00014994136807817588,
+      "loss": 0.4228,
+      "step": 3940
+    },
+    {
+      "epoch": 2.54,
+      "learning_rate": 0.0001495504885993485,
+      "loss": 0.3963,
+      "step": 3950
+    },
+    {
+      "epoch": 2.55,
+      "learning_rate": 0.00014915960912052117,
+      "loss": 0.4132,
+      "step": 3960
+    },
+    {
+      "epoch": 2.55,
+      "learning_rate": 0.0001487687296416938,
+      "loss": 0.4273,
+      "step": 3970
+    },
+    {
+      "epoch": 2.56,
+      "learning_rate": 0.00014837785016286642,
+      "loss": 0.4153,
+      "step": 3980
+    },
+    {
+      "epoch": 2.57,
+      "learning_rate": 0.00014798697068403908,
+      "loss": 0.4379,
+      "step": 3990
+    },
+    {
+      "epoch": 2.57,
+      "learning_rate": 0.0001475960912052117,
+      "loss": 0.4195,
+      "step": 4000
+    },
+    {
+      "epoch": 2.57,
+      "eval_loss": 0.4522078335285187,
+      "eval_runtime": 403.1528,
+      "eval_samples_per_second": 4.961,
+      "eval_steps_per_second": 0.62,
+      "step": 4000
+    },
+    {
+      "epoch": 2.58,
+      "learning_rate": 0.00014720521172638436,
+      "loss": 0.4242,
+      "step": 4010
+    },
+    {
+      "epoch": 2.59,
+      "learning_rate": 0.000146814332247557,
+      "loss": 0.4173,
+      "step": 4020
+    },
+    {
+      "epoch": 2.59,
+      "learning_rate": 0.00014642345276872962,
+      "loss": 0.4152,
+      "step": 4030
+    },
+    {
+      "epoch": 2.6,
+      "learning_rate": 0.00014603257328990227,
+      "loss": 0.4143,
+      "step": 4040
+    },
+    {
+      "epoch": 2.6,
+      "learning_rate": 0.0001456416938110749,
+      "loss": 0.4148,
+      "step": 4050
+    },
+    {
+      "epoch": 2.61,
+      "learning_rate": 0.00014525081433224756,
+      "loss": 0.4129,
+      "step": 4060
+    },
+    {
+      "epoch": 2.62,
+      "learning_rate": 0.00014485993485342018,
+      "loss": 0.4065,
+      "step": 4070
+    },
+    {
+      "epoch": 2.62,
+      "learning_rate": 0.0001444690553745928,
+      "loss": 0.4164,
+      "step": 4080
+    },
+    {
+      "epoch": 2.63,
+      "learning_rate": 0.00014407817589576547,
+      "loss": 0.4224,
+      "step": 4090
+    },
+    {
+      "epoch": 2.64,
+      "learning_rate": 0.0001436872964169381,
+      "loss": 0.4303,
+      "step": 4100
+    },
+    {
+      "epoch": 2.64,
+      "learning_rate": 0.00014329641693811072,
+      "loss": 0.4219,
+      "step": 4110
+    },
+    {
+      "epoch": 2.65,
+      "learning_rate": 0.00014290553745928338,
+      "loss": 0.4526,
+      "step": 4120
+    },
+    {
+      "epoch": 2.66,
+      "learning_rate": 0.000142514657980456,
+      "loss": 0.4018,
+      "step": 4130
+    },
+    {
+      "epoch": 2.66,
+      "learning_rate": 0.00014212377850162866,
+      "loss": 0.4166,
+      "step": 4140
+    },
+    {
+      "epoch": 2.67,
+      "learning_rate": 0.0001417328990228013,
+      "loss": 0.4121,
+      "step": 4150
+    },
+    {
+      "epoch": 2.68,
+      "learning_rate": 0.00014134201954397392,
+      "loss": 0.4295,
+      "step": 4160
+    },
+    {
+      "epoch": 2.68,
+      "learning_rate": 0.00014095114006514657,
+      "loss": 0.4093,
+      "step": 4170
+    },
+    {
+      "epoch": 2.69,
+      "learning_rate": 0.0001405602605863192,
+      "loss": 0.423,
+      "step": 4180
+    },
+    {
+      "epoch": 2.69,
+      "learning_rate": 0.00014016938110749186,
+      "loss": 0.4403,
+      "step": 4190
+    },
+    {
+      "epoch": 2.7,
+      "learning_rate": 0.00013977850162866449,
+      "loss": 0.419,
+      "step": 4200
+    },
+    {
+      "epoch": 2.7,
+      "eval_loss": 0.45203742384910583,
+      "eval_runtime": 217.2799,
+      "eval_samples_per_second": 9.205,
+      "eval_steps_per_second": 1.151,
+      "step": 4200
+    },
+    {
+      "epoch": 2.71,
+      "learning_rate": 0.0001393876221498371,
+      "loss": 0.4119,
+      "step": 4210
+    },
+    {
+      "epoch": 2.71,
+      "learning_rate": 0.00013899674267100977,
+      "loss": 0.4062,
+      "step": 4220
+    },
+    {
+      "epoch": 2.72,
+      "learning_rate": 0.0001386058631921824,
+      "loss": 0.4231,
+      "step": 4230
+    },
+    {
+      "epoch": 2.73,
+      "learning_rate": 0.00013821498371335502,
+      "loss": 0.4285,
+      "step": 4240
+    },
+    {
+      "epoch": 2.73,
+      "learning_rate": 0.00013782410423452768,
+      "loss": 0.4026,
+      "step": 4250
+    },
+    {
+      "epoch": 2.74,
+      "learning_rate": 0.0001374332247557003,
+      "loss": 0.4212,
+      "step": 4260
+    },
+    {
+      "epoch": 2.75,
+      "learning_rate": 0.00013704234527687296,
+      "loss": 0.4314,
+      "step": 4270
+    },
+    {
+      "epoch": 2.75,
+      "learning_rate": 0.0001366514657980456,
+      "loss": 0.4181,
+      "step": 4280
+    },
+    {
+      "epoch": 2.76,
+      "learning_rate": 0.00013626058631921822,
+      "loss": 0.4077,
+      "step": 4290
+    },
+    {
+      "epoch": 2.77,
+      "learning_rate": 0.00013586970684039087,
+      "loss": 0.4252,
+      "step": 4300
+    },
+    {
+      "epoch": 2.77,
+      "learning_rate": 0.0001354788273615635,
+      "loss": 0.4166,
+      "step": 4310
+    },
+    {
+      "epoch": 2.78,
+      "learning_rate": 0.00013508794788273616,
+      "loss": 0.4277,
+      "step": 4320
+    },
+    {
+      "epoch": 2.78,
+      "learning_rate": 0.0001346970684039088,
+      "loss": 0.4255,
+      "step": 4330
+    },
+    {
+      "epoch": 2.79,
+      "learning_rate": 0.00013430618892508141,
+      "loss": 0.4257,
+      "step": 4340
+    },
+    {
+      "epoch": 2.8,
+      "learning_rate": 0.00013391530944625407,
+      "loss": 0.4236,
+      "step": 4350
+    },
+    {
+      "epoch": 2.8,
+      "learning_rate": 0.0001335244299674267,
+      "loss": 0.411,
+      "step": 4360
+    },
+    {
+      "epoch": 2.81,
+      "learning_rate": 0.00013313355048859933,
+      "loss": 0.4317,
+      "step": 4370
+    },
+    {
+      "epoch": 2.82,
+      "learning_rate": 0.00013274267100977198,
+      "loss": 0.4188,
+      "step": 4380
+    },
+    {
+      "epoch": 2.82,
+      "learning_rate": 0.0001323517915309446,
+      "loss": 0.4295,
+      "step": 4390
+    },
+    {
+      "epoch": 2.83,
+      "learning_rate": 0.00013196091205211726,
+      "loss": 0.4345,
+      "step": 4400
+    },
+    {
+      "epoch": 2.83,
+      "eval_loss": 0.4510750472545624,
+      "eval_runtime": 217.2642,
+      "eval_samples_per_second": 9.205,
+      "eval_steps_per_second": 1.151,
+      "step": 4400
+    },
+    {
+      "epoch": 2.84,
+      "learning_rate": 0.0001315700325732899,
+      "loss": 0.4256,
+      "step": 4410
+    },
+    {
+      "epoch": 2.84,
+      "learning_rate": 0.00013117915309446252,
+      "loss": 0.4269,
+      "step": 4420
+    },
+    {
+      "epoch": 2.85,
+      "learning_rate": 0.00013078827361563518,
+      "loss": 0.431,
+      "step": 4430
+    },
+    {
+      "epoch": 2.86,
+      "learning_rate": 0.0001303973941368078,
+      "loss": 0.4019,
+      "step": 4440
+    },
+    {
+      "epoch": 2.86,
+      "learning_rate": 0.00013000651465798046,
+      "loss": 0.431,
+      "step": 4450
+    },
+    {
+      "epoch": 2.87,
+      "learning_rate": 0.0001296156351791531,
+      "loss": 0.4039,
+      "step": 4460
+    },
+    {
+      "epoch": 2.87,
+      "learning_rate": 0.00012922475570032572,
+      "loss": 0.4099,
+      "step": 4470
+    },
+    {
+      "epoch": 2.88,
+      "learning_rate": 0.00012883387622149837,
+      "loss": 0.3961,
+      "step": 4480
+    },
+    {
+      "epoch": 2.89,
+      "learning_rate": 0.000128442996742671,
+      "loss": 0.4312,
+      "step": 4490
+    },
+    {
+      "epoch": 2.89,
+      "learning_rate": 0.00012805211726384363,
+      "loss": 0.4248,
+      "step": 4500
+    },
+    {
+      "epoch": 2.9,
+      "learning_rate": 0.00012766123778501626,
+      "loss": 0.4114,
+      "step": 4510
+    },
+    {
+      "epoch": 2.91,
+      "learning_rate": 0.0001272703583061889,
+      "loss": 0.4213,
+      "step": 4520
+    },
+    {
+      "epoch": 2.91,
+      "learning_rate": 0.00012687947882736157,
+      "loss": 0.4118,
+      "step": 4530
+    },
+    {
+      "epoch": 2.92,
+      "learning_rate": 0.0001264885993485342,
+      "loss": 0.4281,
+      "step": 4540
+    },
+    {
+      "epoch": 2.93,
+      "learning_rate": 0.00012609771986970682,
+      "loss": 0.4116,
+      "step": 4550
+    },
+    {
+      "epoch": 2.93,
+      "learning_rate": 0.00012570684039087945,
+      "loss": 0.4311,
+      "step": 4560
+    },
+    {
+      "epoch": 2.94,
+      "learning_rate": 0.0001253159609120521,
+      "loss": 0.4187,
+      "step": 4570
+    },
+    {
+      "epoch": 2.95,
+      "learning_rate": 0.00012492508143322476,
+      "loss": 0.4101,
+      "step": 4580
+    },
+    {
+      "epoch": 2.95,
+      "learning_rate": 0.0001245342019543974,
+      "loss": 0.4029,
+      "step": 4590
+    },
+    {
+      "epoch": 2.96,
+      "learning_rate": 0.00012414332247557002,
+      "loss": 0.4211,
+      "step": 4600
+    },
+    {
+      "epoch": 2.96,
+      "eval_loss": 0.4508701264858246,
+      "eval_runtime": 218.851,
+      "eval_samples_per_second": 9.139,
+      "eval_steps_per_second": 1.142,
+      "step": 4600
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 7775,
+  "num_train_epochs": 5,
+  "save_steps": 200,
+  "total_flos": 2.3380063406159954e+18,
+  "trial_name": null,
+  "trial_params": null
+}

with_input/1b_52k/best_model/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a596f2d7e468432593e0605737d3cff6d7ba32b0ca82291a3251c26b06f9230b
+size 4027

with_input/decomp_code/best_model/adapter_config.json

@@ -0,0 +1,21 @@
+{
+  "base_model_name_or_path": "huggyllama/llama-7b",
+  "bias": "none",
+  "enable_lora": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "lora_alpha": 16,
+  "lora_dropout": 0.05,
+  "merge_weights": false,
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 16,
+  "target_modules": [
+    "q_proj",
+    "k_proj",
+    "v_proj",
+    "o_proj"
+  ],
+  "task_type": "CAUSAL_LM"
+}

with_input/decomp_code/best_model/adapter_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1b4ebf561292631833be935c6693945069d8479d27d8bb5137dde66dd9e42905
+size 67201357

with_input/decomp_code/best_model/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e3bff96a47e6226b0b03bab89b2ae1c35ddeb164a4f0244eb82ec9082e0f4e93
+size 134433093

with_input/decomp_code/best_model/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:86f858b23b33ce0cc15b9fbc31247e8c7c21dbacd7da0b1da35c1c5b2e91fb21
+size 14575

with_input/decomp_code/best_model/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:aaf3b274bb9cc118ee254a99daf710e7721c99ca8673122b1053fda825177be7
+size 627

with_input/decomp_code/best_model/trainer_state.json

@@ -0,0 +1,2963 @@
+{
+  "best_metric": 0.35899612307548523,
+  "best_model_checkpoint": "checkpoints/instrucode/with_input/decomp_code/checkpoint-4600",
+  "epoch": 2.958199356913183,
+  "eval_steps": 200,
+  "global_step": 4600,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.01,
+      "learning_rate": 2.9999999999999997e-05,
+      "loss": 1.2434,
+      "step": 10
+    },
+    {
+      "epoch": 0.01,
+      "learning_rate": 5.9999999999999995e-05,
+      "loss": 1.1915,
+      "step": 20
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 8.999999999999999e-05,
+      "loss": 1.0439,
+      "step": 30
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 0.00011999999999999999,
+      "loss": 0.7455,
+      "step": 40
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 0.00015,
+      "loss": 0.6043,
+      "step": 50
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 0.00017999999999999998,
+      "loss": 0.5039,
+      "step": 60
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 0.00020999999999999998,
+      "loss": 0.4772,
+      "step": 70
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 0.00023999999999999998,
+      "loss": 0.4685,
+      "step": 80
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 0.00027,
+      "loss": 0.4364,
+      "step": 90
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 0.0003,
+      "loss": 0.4473,
+      "step": 100
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 0.00029960912052117263,
+      "loss": 0.4525,
+      "step": 110
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 0.0002992182410423453,
+      "loss": 0.4295,
+      "step": 120
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 0.0002988273615635179,
+      "loss": 0.432,
+      "step": 130
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 0.00029843648208469054,
+      "loss": 0.4316,
+      "step": 140
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 0.00029804560260586314,
+      "loss": 0.4317,
+      "step": 150
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 0.0002976547231270358,
+      "loss": 0.4214,
+      "step": 160
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 0.00029726384364820845,
+      "loss": 0.4127,
+      "step": 170
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 0.00029687296416938105,
+      "loss": 0.431,
+      "step": 180
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 0.0002964820846905537,
+      "loss": 0.4194,
+      "step": 190
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 0.00029609120521172636,
+      "loss": 0.4155,
+      "step": 200
+    },
+    {
+      "epoch": 0.13,
+      "eval_loss": 0.4197882413864136,
+      "eval_runtime": 264.5442,
+      "eval_samples_per_second": 7.56,
+      "eval_steps_per_second": 0.945,
+      "step": 200
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 0.000295700325732899,
+      "loss": 0.4023,
+      "step": 210
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 0.0002953094462540716,
+      "loss": 0.419,
+      "step": 220
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 0.0002949185667752443,
+      "loss": 0.4138,
+      "step": 230
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 0.00029452768729641693,
+      "loss": 0.4143,
+      "step": 240
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 0.00029413680781758953,
+      "loss": 0.4135,
+      "step": 250
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 0.0002937459283387622,
+      "loss": 0.4162,
+      "step": 260
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 0.00029335504885993484,
+      "loss": 0.4236,
+      "step": 270
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 0.00029296416938110744,
+      "loss": 0.4051,
+      "step": 280
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 0.0002925732899022801,
+      "loss": 0.4008,
+      "step": 290
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 0.00029218241042345275,
+      "loss": 0.4169,
+      "step": 300
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 0.0002917915309446254,
+      "loss": 0.4101,
+      "step": 310
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 0.000291400651465798,
+      "loss": 0.3928,
+      "step": 320
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 0.00029100977198697066,
+      "loss": 0.4064,
+      "step": 330
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 0.00029061889250814327,
+      "loss": 0.3884,
+      "step": 340
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 0.0002902280130293159,
+      "loss": 0.3883,
+      "step": 350
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 0.0002898371335504886,
+      "loss": 0.4054,
+      "step": 360
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 0.00028944625407166123,
+      "loss": 0.3893,
+      "step": 370
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 0.0002890553745928339,
+      "loss": 0.3931,
+      "step": 380
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 0.0002886644951140065,
+      "loss": 0.4085,
+      "step": 390
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 0.00028827361563517914,
+      "loss": 0.4082,
+      "step": 400
+    },
+    {
+      "epoch": 0.26,
+      "eval_loss": 0.40093666315078735,
+      "eval_runtime": 264.5209,
+      "eval_samples_per_second": 7.561,
+      "eval_steps_per_second": 0.945,
+      "step": 400
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 0.00028788273615635174,
+      "loss": 0.391,
+      "step": 410
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 0.0002874918566775244,
+      "loss": 0.3815,
+      "step": 420
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 0.00028710097719869705,
+      "loss": 0.3916,
+      "step": 430
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 0.00028671009771986966,
+      "loss": 0.4017,
+      "step": 440
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 0.0002863192182410423,
+      "loss": 0.3946,
+      "step": 450
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 0.00028592833876221497,
+      "loss": 0.3951,
+      "step": 460
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 0.0002855374592833876,
+      "loss": 0.3992,
+      "step": 470
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 0.0002851465798045602,
+      "loss": 0.3834,
+      "step": 480
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 0.0002847557003257329,
+      "loss": 0.4017,
+      "step": 490
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 0.0002843648208469055,
+      "loss": 0.3898,
+      "step": 500
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 0.00028397394136807813,
+      "loss": 0.3882,
+      "step": 510
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 0.0002835830618892508,
+      "loss": 0.3946,
+      "step": 520
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 0.00028319218241042344,
+      "loss": 0.392,
+      "step": 530
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 0.00028280130293159605,
+      "loss": 0.3939,
+      "step": 540
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 0.0002824104234527687,
+      "loss": 0.3805,
+      "step": 550
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 0.00028201954397394136,
+      "loss": 0.3887,
+      "step": 560
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 0.000281628664495114,
+      "loss": 0.3922,
+      "step": 570
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 0.0002812377850162866,
+      "loss": 0.3804,
+      "step": 580
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 0.00028084690553745927,
+      "loss": 0.3943,
+      "step": 590
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 0.00028045602605863187,
+      "loss": 0.3843,
+      "step": 600
+    },
+    {
+      "epoch": 0.39,
+      "eval_loss": 0.3920394778251648,
+      "eval_runtime": 264.4278,
+      "eval_samples_per_second": 7.564,
+      "eval_steps_per_second": 0.945,
+      "step": 600
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 0.0002800651465798045,
+      "loss": 0.3828,
+      "step": 610
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 0.0002796742671009772,
+      "loss": 0.3824,
+      "step": 620
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 0.00027928338762214983,
+      "loss": 0.3902,
+      "step": 630
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 0.00027889250814332244,
+      "loss": 0.3901,
+      "step": 640
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 0.0002785016286644951,
+      "loss": 0.3759,
+      "step": 650
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 0.00027811074918566775,
+      "loss": 0.3849,
+      "step": 660
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 0.00027771986970684035,
+      "loss": 0.3893,
+      "step": 670
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 0.000277328990228013,
+      "loss": 0.3792,
+      "step": 680
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 0.00027693811074918566,
+      "loss": 0.3807,
+      "step": 690
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 0.00027654723127035826,
+      "loss": 0.3864,
+      "step": 700
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 0.0002761563517915309,
+      "loss": 0.3743,
+      "step": 710
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 0.00027576547231270357,
+      "loss": 0.3712,
+      "step": 720
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 0.0002753745928338762,
+      "loss": 0.3794,
+      "step": 730
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 0.0002749837133550488,
+      "loss": 0.3865,
+      "step": 740
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 0.0002745928338762215,
+      "loss": 0.3823,
+      "step": 750
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 0.0002742019543973941,
+      "loss": 0.3698,
+      "step": 760
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 0.00027381107491856674,
+      "loss": 0.3837,
+      "step": 770
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 0.0002734201954397394,
+      "loss": 0.378,
+      "step": 780
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 0.00027302931596091205,
+      "loss": 0.3884,
+      "step": 790
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 0.00027263843648208465,
+      "loss": 0.3953,
+      "step": 800
+    },
+    {
+      "epoch": 0.51,
+      "eval_loss": 0.3857077658176422,
+      "eval_runtime": 264.3754,
+      "eval_samples_per_second": 7.565,
+      "eval_steps_per_second": 0.946,
+      "step": 800
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 0.0002722475570032573,
+      "loss": 0.3722,
+      "step": 810
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 0.00027185667752442996,
+      "loss": 0.3821,
+      "step": 820
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 0.0002714657980456026,
+      "loss": 0.3641,
+      "step": 830
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 0.0002710749185667752,
+      "loss": 0.3785,
+      "step": 840
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 0.00027068403908794787,
+      "loss": 0.3778,
+      "step": 850
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 0.00027029315960912047,
+      "loss": 0.3843,
+      "step": 860
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 0.0002699022801302931,
+      "loss": 0.3961,
+      "step": 870
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 0.0002695114006514658,
+      "loss": 0.3741,
+      "step": 880
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 0.00026912052117263844,
+      "loss": 0.38,
+      "step": 890
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 0.00026872964169381104,
+      "loss": 0.3838,
+      "step": 900
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 0.0002683387622149837,
+      "loss": 0.3739,
+      "step": 910
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 0.00026794788273615635,
+      "loss": 0.3817,
+      "step": 920
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 0.00026755700325732895,
+      "loss": 0.3868,
+      "step": 930
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 0.0002671661237785016,
+      "loss": 0.3722,
+      "step": 940
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 0.0002667752442996742,
+      "loss": 0.3981,
+      "step": 950
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 0.00026638436482084686,
+      "loss": 0.3715,
+      "step": 960
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 0.0002659934853420195,
+      "loss": 0.3832,
+      "step": 970
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 0.00026560260586319217,
+      "loss": 0.3666,
+      "step": 980
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 0.0002652117263843648,
+      "loss": 0.391,
+      "step": 990
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 0.00026482084690553743,
+      "loss": 0.3862,
+      "step": 1000
+    },
+    {
+      "epoch": 0.64,
+      "eval_loss": 0.3816088140010834,
+      "eval_runtime": 264.3954,
+      "eval_samples_per_second": 7.564,
+      "eval_steps_per_second": 0.946,
+      "step": 1000
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 0.0002644299674267101,
+      "loss": 0.3866,
+      "step": 1010
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 0.0002640390879478827,
+      "loss": 0.3802,
+      "step": 1020
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 0.00026364820846905534,
+      "loss": 0.3834,
+      "step": 1030
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 0.000263257328990228,
+      "loss": 0.376,
+      "step": 1040
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 0.0002628664495114006,
+      "loss": 0.3678,
+      "step": 1050
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 0.00026247557003257325,
+      "loss": 0.3736,
+      "step": 1060
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 0.0002620846905537459,
+      "loss": 0.3698,
+      "step": 1070
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 0.00026169381107491856,
+      "loss": 0.3744,
+      "step": 1080
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 0.0002613029315960912,
+      "loss": 0.367,
+      "step": 1090
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 0.0002609120521172638,
+      "loss": 0.3658,
+      "step": 1100
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 0.0002605211726384365,
+      "loss": 0.3662,
+      "step": 1110
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 0.0002601302931596091,
+      "loss": 0.3839,
+      "step": 1120
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 0.00025973941368078173,
+      "loss": 0.3756,
+      "step": 1130
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 0.0002593485342019544,
+      "loss": 0.3681,
+      "step": 1140
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 0.00025895765472312704,
+      "loss": 0.3843,
+      "step": 1150
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 0.00025856677524429964,
+      "loss": 0.3691,
+      "step": 1160
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 0.0002581758957654723,
+      "loss": 0.3599,
+      "step": 1170
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 0.00025778501628664495,
+      "loss": 0.3758,
+      "step": 1180
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 0.00025739413680781755,
+      "loss": 0.3868,
+      "step": 1190
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 0.0002570032573289902,
+      "loss": 0.3737,
+      "step": 1200
+    },
+    {
+      "epoch": 0.77,
+      "eval_loss": 0.3774874210357666,
+      "eval_runtime": 264.7082,
+      "eval_samples_per_second": 7.555,
+      "eval_steps_per_second": 0.944,
+      "step": 1200
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 0.0002566123778501628,
+      "loss": 0.3676,
+      "step": 1210
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 0.00025622149837133546,
+      "loss": 0.3793,
+      "step": 1220
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 0.0002558306188925081,
+      "loss": 0.3657,
+      "step": 1230
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 0.0002554397394136808,
+      "loss": 0.3657,
+      "step": 1240
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 0.00025504885993485343,
+      "loss": 0.3774,
+      "step": 1250
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 0.00025465798045602603,
+      "loss": 0.3789,
+      "step": 1260
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 0.0002542671009771987,
+      "loss": 0.3609,
+      "step": 1270
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 0.0002538762214983713,
+      "loss": 0.3751,
+      "step": 1280
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 0.00025348534201954394,
+      "loss": 0.381,
+      "step": 1290
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 0.0002530944625407166,
+      "loss": 0.3605,
+      "step": 1300
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 0.0002527035830618892,
+      "loss": 0.3631,
+      "step": 1310
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 0.00025231270358306185,
+      "loss": 0.3643,
+      "step": 1320
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 0.0002519218241042345,
+      "loss": 0.3803,
+      "step": 1330
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 0.00025153094462540716,
+      "loss": 0.3845,
+      "step": 1340
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 0.0002511400651465798,
+      "loss": 0.3779,
+      "step": 1350
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 0.0002507491856677524,
+      "loss": 0.3804,
+      "step": 1360
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 0.0002503583061889251,
+      "loss": 0.3764,
+      "step": 1370
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 0.0002499674267100977,
+      "loss": 0.3704,
+      "step": 1380
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 0.00024957654723127033,
+      "loss": 0.3883,
+      "step": 1390
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 0.000249185667752443,
+      "loss": 0.3791,
+      "step": 1400
+    },
+    {
+      "epoch": 0.9,
+      "eval_loss": 0.37548336386680603,
+      "eval_runtime": 264.4219,
+      "eval_samples_per_second": 7.564,
+      "eval_steps_per_second": 0.945,
+      "step": 1400
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 0.0002487947882736156,
+      "loss": 0.379,
+      "step": 1410
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 0.00024840390879478824,
+      "loss": 0.3706,
+      "step": 1420
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 0.0002480130293159609,
+      "loss": 0.3672,
+      "step": 1430
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 0.00024762214983713355,
+      "loss": 0.3693,
+      "step": 1440
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 0.00024723127035830616,
+      "loss": 0.3751,
+      "step": 1450
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 0.0002468403908794788,
+      "loss": 0.3839,
+      "step": 1460
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 0.0002464495114006514,
+      "loss": 0.3785,
+      "step": 1470
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 0.00024605863192182407,
+      "loss": 0.3693,
+      "step": 1480
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 0.0002456677524429967,
+      "loss": 0.3697,
+      "step": 1490
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 0.0002452768729641694,
+      "loss": 0.3722,
+      "step": 1500
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 0.00024488599348534203,
+      "loss": 0.3705,
+      "step": 1510
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 0.00024449511400651463,
+      "loss": 0.3608,
+      "step": 1520
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 0.00024410423452768726,
+      "loss": 0.3652,
+      "step": 1530
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 0.00024371335504885992,
+      "loss": 0.371,
+      "step": 1540
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 0.00024332247557003257,
+      "loss": 0.364,
+      "step": 1550
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 0.0002429315960912052,
+      "loss": 0.3557,
+      "step": 1560
+    },
+    {
+      "epoch": 1.01,
+      "learning_rate": 0.00024254071661237783,
+      "loss": 0.363,
+      "step": 1570
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 0.00024214983713355046,
+      "loss": 0.3593,
+      "step": 1580
+    },
+    {
+      "epoch": 1.02,
+      "learning_rate": 0.0002417589576547231,
+      "loss": 0.3668,
+      "step": 1590
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 0.00024136807817589574,
+      "loss": 0.3562,
+      "step": 1600
+    },
+    {
+      "epoch": 1.03,
+      "eval_loss": 0.3727501332759857,
+      "eval_runtime": 264.8173,
+      "eval_samples_per_second": 7.552,
+      "eval_steps_per_second": 0.944,
+      "step": 1600
+    },
+    {
+      "epoch": 1.04,
+      "learning_rate": 0.0002409771986970684,
+      "loss": 0.3628,
+      "step": 1610
+    },
+    {
+      "epoch": 1.04,
+      "learning_rate": 0.000240586319218241,
+      "loss": 0.3658,
+      "step": 1620
+    },
+    {
+      "epoch": 1.05,
+      "learning_rate": 0.00024019543973941365,
+      "loss": 0.3393,
+      "step": 1630
+    },
+    {
+      "epoch": 1.05,
+      "learning_rate": 0.0002398045602605863,
+      "loss": 0.3567,
+      "step": 1640
+    },
+    {
+      "epoch": 1.06,
+      "learning_rate": 0.00023941368078175893,
+      "loss": 0.3583,
+      "step": 1650
+    },
+    {
+      "epoch": 1.07,
+      "learning_rate": 0.0002390228013029316,
+      "loss": 0.3636,
+      "step": 1660
+    },
+    {
+      "epoch": 1.07,
+      "learning_rate": 0.0002386319218241042,
+      "loss": 0.3537,
+      "step": 1670
+    },
+    {
+      "epoch": 1.08,
+      "learning_rate": 0.00023824104234527685,
+      "loss": 0.3625,
+      "step": 1680
+    },
+    {
+      "epoch": 1.09,
+      "learning_rate": 0.0002378501628664495,
+      "loss": 0.3575,
+      "step": 1690
+    },
+    {
+      "epoch": 1.09,
+      "learning_rate": 0.00023745928338762213,
+      "loss": 0.3483,
+      "step": 1700
+    },
+    {
+      "epoch": 1.1,
+      "learning_rate": 0.00023706840390879478,
+      "loss": 0.3606,
+      "step": 1710
+    },
+    {
+      "epoch": 1.11,
+      "learning_rate": 0.00023667752442996739,
+      "loss": 0.3595,
+      "step": 1720
+    },
+    {
+      "epoch": 1.11,
+      "learning_rate": 0.00023628664495114004,
+      "loss": 0.3549,
+      "step": 1730
+    },
+    {
+      "epoch": 1.12,
+      "learning_rate": 0.00023589576547231267,
+      "loss": 0.3431,
+      "step": 1740
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 0.00023550488599348532,
+      "loss": 0.3493,
+      "step": 1750
+    },
+    {
+      "epoch": 1.13,
+      "learning_rate": 0.00023511400651465798,
+      "loss": 0.3654,
+      "step": 1760
+    },
+    {
+      "epoch": 1.14,
+      "learning_rate": 0.00023472312703583058,
+      "loss": 0.3601,
+      "step": 1770
+    },
+    {
+      "epoch": 1.14,
+      "learning_rate": 0.00023433224755700324,
+      "loss": 0.3635,
+      "step": 1780
+    },
+    {
+      "epoch": 1.15,
+      "learning_rate": 0.00023394136807817586,
+      "loss": 0.3606,
+      "step": 1790
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 0.00023355048859934852,
+      "loss": 0.3423,
+      "step": 1800
+    },
+    {
+      "epoch": 1.16,
+      "eval_loss": 0.37143847346305847,
+      "eval_runtime": 264.7435,
+      "eval_samples_per_second": 7.554,
+      "eval_steps_per_second": 0.944,
+      "step": 1800
+    },
+    {
+      "epoch": 1.16,
+      "learning_rate": 0.00023315960912052117,
+      "loss": 0.3607,
+      "step": 1810
+    },
+    {
+      "epoch": 1.17,
+      "learning_rate": 0.00023276872964169378,
+      "loss": 0.3573,
+      "step": 1820
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 0.00023237785016286643,
+      "loss": 0.356,
+      "step": 1830
+    },
+    {
+      "epoch": 1.18,
+      "learning_rate": 0.00023198697068403906,
+      "loss": 0.3615,
+      "step": 1840
+    },
+    {
+      "epoch": 1.19,
+      "learning_rate": 0.00023159609120521171,
+      "loss": 0.351,
+      "step": 1850
+    },
+    {
+      "epoch": 1.2,
+      "learning_rate": 0.00023120521172638434,
+      "loss": 0.3486,
+      "step": 1860
+    },
+    {
+      "epoch": 1.2,
+      "learning_rate": 0.000230814332247557,
+      "loss": 0.3651,
+      "step": 1870
+    },
+    {
+      "epoch": 1.21,
+      "learning_rate": 0.0002304234527687296,
+      "loss": 0.354,
+      "step": 1880
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 0.00023003257328990225,
+      "loss": 0.3495,
+      "step": 1890
+    },
+    {
+      "epoch": 1.22,
+      "learning_rate": 0.0002296416938110749,
+      "loss": 0.3625,
+      "step": 1900
+    },
+    {
+      "epoch": 1.23,
+      "learning_rate": 0.00022925081433224754,
+      "loss": 0.3781,
+      "step": 1910
+    },
+    {
+      "epoch": 1.23,
+      "learning_rate": 0.0002288599348534202,
+      "loss": 0.3441,
+      "step": 1920
+    },
+    {
+      "epoch": 1.24,
+      "learning_rate": 0.0002284690553745928,
+      "loss": 0.355,
+      "step": 1930
+    },
+    {
+      "epoch": 1.25,
+      "learning_rate": 0.00022807817589576545,
+      "loss": 0.3577,
+      "step": 1940
+    },
+    {
+      "epoch": 1.25,
+      "learning_rate": 0.0002276872964169381,
+      "loss": 0.3581,
+      "step": 1950
+    },
+    {
+      "epoch": 1.26,
+      "learning_rate": 0.00022729641693811073,
+      "loss": 0.3485,
+      "step": 1960
+    },
+    {
+      "epoch": 1.27,
+      "learning_rate": 0.0002269055374592834,
+      "loss": 0.3503,
+      "step": 1970
+    },
+    {
+      "epoch": 1.27,
+      "learning_rate": 0.000226514657980456,
+      "loss": 0.3415,
+      "step": 1980
+    },
+    {
+      "epoch": 1.28,
+      "learning_rate": 0.00022612377850162864,
+      "loss": 0.3504,
+      "step": 1990
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 0.00022573289902280127,
+      "loss": 0.3542,
+      "step": 2000
+    },
+    {
+      "epoch": 1.29,
+      "eval_loss": 0.37012627720832825,
+      "eval_runtime": 264.6224,
+      "eval_samples_per_second": 7.558,
+      "eval_steps_per_second": 0.945,
+      "step": 2000
+    },
+    {
+      "epoch": 1.29,
+      "learning_rate": 0.00022534201954397393,
+      "loss": 0.351,
+      "step": 2010
+    },
+    {
+      "epoch": 1.3,
+      "learning_rate": 0.00022495114006514658,
+      "loss": 0.3529,
+      "step": 2020
+    },
+    {
+      "epoch": 1.31,
+      "learning_rate": 0.00022456026058631918,
+      "loss": 0.3496,
+      "step": 2030
+    },
+    {
+      "epoch": 1.31,
+      "learning_rate": 0.00022416938110749184,
+      "loss": 0.3639,
+      "step": 2040
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 0.00022377850162866447,
+      "loss": 0.3601,
+      "step": 2050
+    },
+    {
+      "epoch": 1.32,
+      "learning_rate": 0.00022338762214983712,
+      "loss": 0.3436,
+      "step": 2060
+    },
+    {
+      "epoch": 1.33,
+      "learning_rate": 0.00022299674267100978,
+      "loss": 0.3519,
+      "step": 2070
+    },
+    {
+      "epoch": 1.34,
+      "learning_rate": 0.00022260586319218238,
+      "loss": 0.3627,
+      "step": 2080
+    },
+    {
+      "epoch": 1.34,
+      "learning_rate": 0.00022221498371335503,
+      "loss": 0.3485,
+      "step": 2090
+    },
+    {
+      "epoch": 1.35,
+      "learning_rate": 0.00022182410423452766,
+      "loss": 0.348,
+      "step": 2100
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 0.00022143322475570032,
+      "loss": 0.3492,
+      "step": 2110
+    },
+    {
+      "epoch": 1.36,
+      "learning_rate": 0.00022104234527687295,
+      "loss": 0.3433,
+      "step": 2120
+    },
+    {
+      "epoch": 1.37,
+      "learning_rate": 0.00022065146579804557,
+      "loss": 0.3462,
+      "step": 2130
+    },
+    {
+      "epoch": 1.38,
+      "learning_rate": 0.0002202605863192182,
+      "loss": 0.3506,
+      "step": 2140
+    },
+    {
+      "epoch": 1.38,
+      "learning_rate": 0.00021986970684039086,
+      "loss": 0.3407,
+      "step": 2150
+    },
+    {
+      "epoch": 1.39,
+      "learning_rate": 0.0002194788273615635,
+      "loss": 0.3618,
+      "step": 2160
+    },
+    {
+      "epoch": 1.4,
+      "learning_rate": 0.00021908794788273614,
+      "loss": 0.3649,
+      "step": 2170
+    },
+    {
+      "epoch": 1.4,
+      "learning_rate": 0.00021869706840390877,
+      "loss": 0.3411,
+      "step": 2180
+    },
+    {
+      "epoch": 1.41,
+      "learning_rate": 0.0002183061889250814,
+      "loss": 0.3497,
+      "step": 2190
+    },
+    {
+      "epoch": 1.41,
+      "learning_rate": 0.00021791530944625405,
+      "loss": 0.3569,
+      "step": 2200
+    },
+    {
+      "epoch": 1.41,
+      "eval_loss": 0.3684168756008148,
+      "eval_runtime": 264.757,
+      "eval_samples_per_second": 7.554,
+      "eval_steps_per_second": 0.944,
+      "step": 2200
+    },
+    {
+      "epoch": 1.42,
+      "learning_rate": 0.0002175244299674267,
+      "loss": 0.3489,
+      "step": 2210
+    },
+    {
+      "epoch": 1.43,
+      "learning_rate": 0.00021713355048859934,
+      "loss": 0.3538,
+      "step": 2220
+    },
+    {
+      "epoch": 1.43,
+      "learning_rate": 0.000216742671009772,
+      "loss": 0.346,
+      "step": 2230
+    },
+    {
+      "epoch": 1.44,
+      "learning_rate": 0.0002163517915309446,
+      "loss": 0.3708,
+      "step": 2240
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 0.00021596091205211725,
+      "loss": 0.3421,
+      "step": 2250
+    },
+    {
+      "epoch": 1.45,
+      "learning_rate": 0.00021557003257328987,
+      "loss": 0.3456,
+      "step": 2260
+    },
+    {
+      "epoch": 1.46,
+      "learning_rate": 0.00021517915309446253,
+      "loss": 0.3633,
+      "step": 2270
+    },
+    {
+      "epoch": 1.47,
+      "learning_rate": 0.00021478827361563519,
+      "loss": 0.3531,
+      "step": 2280
+    },
+    {
+      "epoch": 1.47,
+      "learning_rate": 0.00021439739413680779,
+      "loss": 0.3553,
+      "step": 2290
+    },
+    {
+      "epoch": 1.48,
+      "learning_rate": 0.00021400651465798044,
+      "loss": 0.3487,
+      "step": 2300
+    },
+    {
+      "epoch": 1.49,
+      "learning_rate": 0.00021361563517915307,
+      "loss": 0.3527,
+      "step": 2310
+    },
+    {
+      "epoch": 1.49,
+      "learning_rate": 0.00021322475570032572,
+      "loss": 0.3489,
+      "step": 2320
+    },
+    {
+      "epoch": 1.5,
+      "learning_rate": 0.00021283387622149838,
+      "loss": 0.3614,
+      "step": 2330
+    },
+    {
+      "epoch": 1.5,
+      "learning_rate": 0.00021244299674267098,
+      "loss": 0.3561,
+      "step": 2340
+    },
+    {
+      "epoch": 1.51,
+      "learning_rate": 0.00021205211726384364,
+      "loss": 0.3617,
+      "step": 2350
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 0.00021166123778501626,
+      "loss": 0.3458,
+      "step": 2360
+    },
+    {
+      "epoch": 1.52,
+      "learning_rate": 0.00021127035830618892,
+      "loss": 0.3555,
+      "step": 2370
+    },
+    {
+      "epoch": 1.53,
+      "learning_rate": 0.00021087947882736155,
+      "loss": 0.3518,
+      "step": 2380
+    },
+    {
+      "epoch": 1.54,
+      "learning_rate": 0.00021048859934853418,
+      "loss": 0.3537,
+      "step": 2390
+    },
+    {
+      "epoch": 1.54,
+      "learning_rate": 0.0002100977198697068,
+      "loss": 0.3525,
+      "step": 2400
+    },
+    {
+      "epoch": 1.54,
+      "eval_loss": 0.3664863109588623,
+      "eval_runtime": 264.8744,
+      "eval_samples_per_second": 7.551,
+      "eval_steps_per_second": 0.944,
+      "step": 2400
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 0.00020970684039087946,
+      "loss": 0.3487,
+      "step": 2410
+    },
+    {
+      "epoch": 1.56,
+      "learning_rate": 0.00020931596091205211,
+      "loss": 0.3525,
+      "step": 2420
+    },
+    {
+      "epoch": 1.56,
+      "learning_rate": 0.00020892508143322474,
+      "loss": 0.3504,
+      "step": 2430
+    },
+    {
+      "epoch": 1.57,
+      "learning_rate": 0.00020853420195439737,
+      "loss": 0.3482,
+      "step": 2440
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 0.00020814332247557,
+      "loss": 0.3592,
+      "step": 2450
+    },
+    {
+      "epoch": 1.58,
+      "learning_rate": 0.00020775244299674265,
+      "loss": 0.3525,
+      "step": 2460
+    },
+    {
+      "epoch": 1.59,
+      "learning_rate": 0.0002073615635179153,
+      "loss": 0.3489,
+      "step": 2470
+    },
+    {
+      "epoch": 1.59,
+      "learning_rate": 0.00020697068403908794,
+      "loss": 0.3597,
+      "step": 2480
+    },
+    {
+      "epoch": 1.6,
+      "learning_rate": 0.00020657980456026057,
+      "loss": 0.3467,
+      "step": 2490
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 0.0002061889250814332,
+      "loss": 0.3622,
+      "step": 2500
+    },
+    {
+      "epoch": 1.61,
+      "learning_rate": 0.00020579804560260585,
+      "loss": 0.3685,
+      "step": 2510
+    },
+    {
+      "epoch": 1.62,
+      "learning_rate": 0.00020540716612377848,
+      "loss": 0.3524,
+      "step": 2520
+    },
+    {
+      "epoch": 1.63,
+      "learning_rate": 0.00020501628664495113,
+      "loss": 0.3378,
+      "step": 2530
+    },
+    {
+      "epoch": 1.63,
+      "learning_rate": 0.00020462540716612373,
+      "loss": 0.3408,
+      "step": 2540
+    },
+    {
+      "epoch": 1.64,
+      "learning_rate": 0.0002042345276872964,
+      "loss": 0.352,
+      "step": 2550
+    },
+    {
+      "epoch": 1.65,
+      "learning_rate": 0.00020384364820846904,
+      "loss": 0.3394,
+      "step": 2560
+    },
+    {
+      "epoch": 1.65,
+      "learning_rate": 0.00020345276872964167,
+      "loss": 0.3491,
+      "step": 2570
+    },
+    {
+      "epoch": 1.66,
+      "learning_rate": 0.00020306188925081433,
+      "loss": 0.3383,
+      "step": 2580
+    },
+    {
+      "epoch": 1.67,
+      "learning_rate": 0.00020267100977198693,
+      "loss": 0.3562,
+      "step": 2590
+    },
+    {
+      "epoch": 1.67,
+      "learning_rate": 0.00020228013029315958,
+      "loss": 0.3542,
+      "step": 2600
+    },
+    {
+      "epoch": 1.67,
+      "eval_loss": 0.36559873819351196,
+      "eval_runtime": 264.7005,
+      "eval_samples_per_second": 7.556,
+      "eval_steps_per_second": 0.944,
+      "step": 2600
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 0.00020188925081433224,
+      "loss": 0.3558,
+      "step": 2610
+    },
+    {
+      "epoch": 1.68,
+      "learning_rate": 0.00020149837133550487,
+      "loss": 0.3619,
+      "step": 2620
+    },
+    {
+      "epoch": 1.69,
+      "learning_rate": 0.00020110749185667752,
+      "loss": 0.3455,
+      "step": 2630
+    },
+    {
+      "epoch": 1.7,
+      "learning_rate": 0.00020071661237785015,
+      "loss": 0.3569,
+      "step": 2640
+    },
+    {
+      "epoch": 1.7,
+      "learning_rate": 0.00020032573289902278,
+      "loss": 0.3441,
+      "step": 2650
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 0.0001999348534201954,
+      "loss": 0.3411,
+      "step": 2660
+    },
+    {
+      "epoch": 1.72,
+      "learning_rate": 0.00019954397394136806,
+      "loss": 0.3577,
+      "step": 2670
+    },
+    {
+      "epoch": 1.72,
+      "learning_rate": 0.00019915309446254072,
+      "loss": 0.3366,
+      "step": 2680
+    },
+    {
+      "epoch": 1.73,
+      "learning_rate": 0.00019876221498371335,
+      "loss": 0.3475,
+      "step": 2690
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 0.00019837133550488597,
+      "loss": 0.3513,
+      "step": 2700
+    },
+    {
+      "epoch": 1.74,
+      "learning_rate": 0.0001979804560260586,
+      "loss": 0.3477,
+      "step": 2710
+    },
+    {
+      "epoch": 1.75,
+      "learning_rate": 0.00019758957654723126,
+      "loss": 0.3509,
+      "step": 2720
+    },
+    {
+      "epoch": 1.76,
+      "learning_rate": 0.0001971986970684039,
+      "loss": 0.3574,
+      "step": 2730
+    },
+    {
+      "epoch": 1.76,
+      "learning_rate": 0.00019680781758957654,
+      "loss": 0.3556,
+      "step": 2740
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 0.00019641693811074917,
+      "loss": 0.3464,
+      "step": 2750
+    },
+    {
+      "epoch": 1.77,
+      "learning_rate": 0.0001960260586319218,
+      "loss": 0.3545,
+      "step": 2760
+    },
+    {
+      "epoch": 1.78,
+      "learning_rate": 0.00019563517915309445,
+      "loss": 0.3618,
+      "step": 2770
+    },
+    {
+      "epoch": 1.79,
+      "learning_rate": 0.00019524429967426708,
+      "loss": 0.3334,
+      "step": 2780
+    },
+    {
+      "epoch": 1.79,
+      "learning_rate": 0.00019485342019543974,
+      "loss": 0.3494,
+      "step": 2790
+    },
+    {
+      "epoch": 1.8,
+      "learning_rate": 0.00019446254071661234,
+      "loss": 0.3575,
+      "step": 2800
+    },
+    {
+      "epoch": 1.8,
+      "eval_loss": 0.3641030788421631,
+      "eval_runtime": 264.7144,
+      "eval_samples_per_second": 7.555,
+      "eval_steps_per_second": 0.944,
+      "step": 2800
+    },
+    {
+      "epoch": 1.81,
+      "learning_rate": 0.000194071661237785,
+      "loss": 0.348,
+      "step": 2810
+    },
+    {
+      "epoch": 1.81,
+      "learning_rate": 0.00019368078175895765,
+      "loss": 0.344,
+      "step": 2820
+    },
+    {
+      "epoch": 1.82,
+      "learning_rate": 0.00019328990228013028,
+      "loss": 0.359,
+      "step": 2830
+    },
+    {
+      "epoch": 1.83,
+      "learning_rate": 0.00019289902280130293,
+      "loss": 0.3596,
+      "step": 2840
+    },
+    {
+      "epoch": 1.83,
+      "learning_rate": 0.00019250814332247553,
+      "loss": 0.3364,
+      "step": 2850
+    },
+    {
+      "epoch": 1.84,
+      "learning_rate": 0.0001921172638436482,
+      "loss": 0.348,
+      "step": 2860
+    },
+    {
+      "epoch": 1.85,
+      "learning_rate": 0.00019172638436482084,
+      "loss": 0.3637,
+      "step": 2870
+    },
+    {
+      "epoch": 1.85,
+      "learning_rate": 0.00019133550488599347,
+      "loss": 0.3474,
+      "step": 2880
+    },
+    {
+      "epoch": 1.86,
+      "learning_rate": 0.00019094462540716613,
+      "loss": 0.3543,
+      "step": 2890
+    },
+    {
+      "epoch": 1.86,
+      "learning_rate": 0.00019055374592833873,
+      "loss": 0.3544,
+      "step": 2900
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 0.00019016286644951138,
+      "loss": 0.3588,
+      "step": 2910
+    },
+    {
+      "epoch": 1.88,
+      "learning_rate": 0.000189771986970684,
+      "loss": 0.355,
+      "step": 2920
+    },
+    {
+      "epoch": 1.88,
+      "learning_rate": 0.00018938110749185666,
+      "loss": 0.3417,
+      "step": 2930
+    },
+    {
+      "epoch": 1.89,
+      "learning_rate": 0.00018899022801302932,
+      "loss": 0.3432,
+      "step": 2940
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 0.00018859934853420192,
+      "loss": 0.3535,
+      "step": 2950
+    },
+    {
+      "epoch": 1.9,
+      "learning_rate": 0.00018820846905537458,
+      "loss": 0.3513,
+      "step": 2960
+    },
+    {
+      "epoch": 1.91,
+      "learning_rate": 0.0001878175895765472,
+      "loss": 0.3556,
+      "step": 2970
+    },
+    {
+      "epoch": 1.92,
+      "learning_rate": 0.00018742671009771986,
+      "loss": 0.3631,
+      "step": 2980
+    },
+    {
+      "epoch": 1.92,
+      "learning_rate": 0.00018703583061889252,
+      "loss": 0.3637,
+      "step": 2990
+    },
+    {
+      "epoch": 1.93,
+      "learning_rate": 0.00018664495114006514,
+      "loss": 0.35,
+      "step": 3000
+    },
+    {
+      "epoch": 1.93,
+      "eval_loss": 0.3628648817539215,
+      "eval_runtime": 264.7329,
+      "eval_samples_per_second": 7.555,
+      "eval_steps_per_second": 0.944,
+      "step": 3000
+    },
+    {
+      "epoch": 1.94,
+      "learning_rate": 0.00018625407166123777,
+      "loss": 0.3573,
+      "step": 3010
+    },
+    {
+      "epoch": 1.94,
+      "learning_rate": 0.0001858631921824104,
+      "loss": 0.3542,
+      "step": 3020
+    },
+    {
+      "epoch": 1.95,
+      "learning_rate": 0.00018547231270358305,
+      "loss": 0.3527,
+      "step": 3030
+    },
+    {
+      "epoch": 1.95,
+      "learning_rate": 0.00018508143322475568,
+      "loss": 0.345,
+      "step": 3040
+    },
+    {
+      "epoch": 1.96,
+      "learning_rate": 0.00018469055374592834,
+      "loss": 0.3353,
+      "step": 3050
+    },
+    {
+      "epoch": 1.97,
+      "learning_rate": 0.00018429967426710094,
+      "loss": 0.3495,
+      "step": 3060
+    },
+    {
+      "epoch": 1.97,
+      "learning_rate": 0.0001839087947882736,
+      "loss": 0.3575,
+      "step": 3070
+    },
+    {
+      "epoch": 1.98,
+      "learning_rate": 0.00018351791530944625,
+      "loss": 0.354,
+      "step": 3080
+    },
+    {
+      "epoch": 1.99,
+      "learning_rate": 0.00018312703583061888,
+      "loss": 0.3337,
+      "step": 3090
+    },
+    {
+      "epoch": 1.99,
+      "learning_rate": 0.00018273615635179153,
+      "loss": 0.3511,
+      "step": 3100
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 0.00018234527687296413,
+      "loss": 0.354,
+      "step": 3110
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 0.0001819543973941368,
+      "loss": 0.3384,
+      "step": 3120
+    },
+    {
+      "epoch": 2.01,
+      "learning_rate": 0.00018156351791530944,
+      "loss": 0.3238,
+      "step": 3130
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 0.00018117263843648207,
+      "loss": 0.3376,
+      "step": 3140
+    },
+    {
+      "epoch": 2.03,
+      "learning_rate": 0.00018078175895765473,
+      "loss": 0.3396,
+      "step": 3150
+    },
+    {
+      "epoch": 2.03,
+      "learning_rate": 0.00018039087947882733,
+      "loss": 0.3397,
+      "step": 3160
+    },
+    {
+      "epoch": 2.04,
+      "learning_rate": 0.00017999999999999998,
+      "loss": 0.3294,
+      "step": 3170
+    },
+    {
+      "epoch": 2.05,
+      "learning_rate": 0.0001796091205211726,
+      "loss": 0.3307,
+      "step": 3180
+    },
+    {
+      "epoch": 2.05,
+      "learning_rate": 0.00017921824104234527,
+      "loss": 0.3226,
+      "step": 3190
+    },
+    {
+      "epoch": 2.06,
+      "learning_rate": 0.00017882736156351792,
+      "loss": 0.3379,
+      "step": 3200
+    },
+    {
+      "epoch": 2.06,
+      "eval_loss": 0.363425612449646,
+      "eval_runtime": 264.6418,
+      "eval_samples_per_second": 7.557,
+      "eval_steps_per_second": 0.945,
+      "step": 3200
+    },
+    {
+      "epoch": 2.06,
+      "learning_rate": 0.00017843648208469052,
+      "loss": 0.3378,
+      "step": 3210
+    },
+    {
+      "epoch": 2.07,
+      "learning_rate": 0.00017804560260586318,
+      "loss": 0.3292,
+      "step": 3220
+    },
+    {
+      "epoch": 2.08,
+      "learning_rate": 0.0001776547231270358,
+      "loss": 0.3421,
+      "step": 3230
+    },
+    {
+      "epoch": 2.08,
+      "learning_rate": 0.00017726384364820846,
+      "loss": 0.3279,
+      "step": 3240
+    },
+    {
+      "epoch": 2.09,
+      "learning_rate": 0.0001768729641693811,
+      "loss": 0.3203,
+      "step": 3250
+    },
+    {
+      "epoch": 2.1,
+      "learning_rate": 0.00017648208469055372,
+      "loss": 0.3448,
+      "step": 3260
+    },
+    {
+      "epoch": 2.1,
+      "learning_rate": 0.00017609120521172637,
+      "loss": 0.3329,
+      "step": 3270
+    },
+    {
+      "epoch": 2.11,
+      "learning_rate": 0.000175700325732899,
+      "loss": 0.3378,
+      "step": 3280
+    },
+    {
+      "epoch": 2.12,
+      "learning_rate": 0.00017530944625407166,
+      "loss": 0.3293,
+      "step": 3290
+    },
+    {
+      "epoch": 2.12,
+      "learning_rate": 0.00017491856677524429,
+      "loss": 0.3455,
+      "step": 3300
+    },
+    {
+      "epoch": 2.13,
+      "learning_rate": 0.00017452768729641691,
+      "loss": 0.3344,
+      "step": 3310
+    },
+    {
+      "epoch": 2.14,
+      "learning_rate": 0.00017413680781758954,
+      "loss": 0.3162,
+      "step": 3320
+    },
+    {
+      "epoch": 2.14,
+      "learning_rate": 0.0001737459283387622,
+      "loss": 0.339,
+      "step": 3330
+    },
+    {
+      "epoch": 2.15,
+      "learning_rate": 0.00017335504885993485,
+      "loss": 0.3348,
+      "step": 3340
+    },
+    {
+      "epoch": 2.15,
+      "learning_rate": 0.00017296416938110748,
+      "loss": 0.3333,
+      "step": 3350
+    },
+    {
+      "epoch": 2.16,
+      "learning_rate": 0.0001725732899022801,
+      "loss": 0.3434,
+      "step": 3360
+    },
+    {
+      "epoch": 2.17,
+      "learning_rate": 0.00017218241042345274,
+      "loss": 0.3307,
+      "step": 3370
+    },
+    {
+      "epoch": 2.17,
+      "learning_rate": 0.0001717915309446254,
+      "loss": 0.3318,
+      "step": 3380
+    },
+    {
+      "epoch": 2.18,
+      "learning_rate": 0.00017140065146579802,
+      "loss": 0.3408,
+      "step": 3390
+    },
+    {
+      "epoch": 2.19,
+      "learning_rate": 0.00017100977198697068,
+      "loss": 0.3465,
+      "step": 3400
+    },
+    {
+      "epoch": 2.19,
+      "eval_loss": 0.3630961775779724,
+      "eval_runtime": 264.7227,
+      "eval_samples_per_second": 7.555,
+      "eval_steps_per_second": 0.944,
+      "step": 3400
+    },
+    {
+      "epoch": 2.19,
+      "learning_rate": 0.00017061889250814333,
+      "loss": 0.3355,
+      "step": 3410
+    },
+    {
+      "epoch": 2.2,
+      "learning_rate": 0.00017022801302931593,
+      "loss": 0.3439,
+      "step": 3420
+    },
+    {
+      "epoch": 2.21,
+      "learning_rate": 0.0001698371335504886,
+      "loss": 0.3269,
+      "step": 3430
+    },
+    {
+      "epoch": 2.21,
+      "learning_rate": 0.00016944625407166122,
+      "loss": 0.3217,
+      "step": 3440
+    },
+    {
+      "epoch": 2.22,
+      "learning_rate": 0.00016905537459283387,
+      "loss": 0.3375,
+      "step": 3450
+    },
+    {
+      "epoch": 2.23,
+      "learning_rate": 0.00016866449511400653,
+      "loss": 0.3396,
+      "step": 3460
+    },
+    {
+      "epoch": 2.23,
+      "learning_rate": 0.00016827361563517913,
+      "loss": 0.3504,
+      "step": 3470
+    },
+    {
+      "epoch": 2.24,
+      "learning_rate": 0.00016788273615635178,
+      "loss": 0.3314,
+      "step": 3480
+    },
+    {
+      "epoch": 2.24,
+      "learning_rate": 0.0001674918566775244,
+      "loss": 0.3362,
+      "step": 3490
+    },
+    {
+      "epoch": 2.25,
+      "learning_rate": 0.00016710097719869707,
+      "loss": 0.3361,
+      "step": 3500
+    },
+    {
+      "epoch": 2.26,
+      "learning_rate": 0.0001667100977198697,
+      "loss": 0.3236,
+      "step": 3510
+    },
+    {
+      "epoch": 2.26,
+      "learning_rate": 0.00016631921824104232,
+      "loss": 0.3229,
+      "step": 3520
+    },
+    {
+      "epoch": 2.27,
+      "learning_rate": 0.00016592833876221495,
+      "loss": 0.3369,
+      "step": 3530
+    },
+    {
+      "epoch": 2.28,
+      "learning_rate": 0.0001655374592833876,
+      "loss": 0.3305,
+      "step": 3540
+    },
+    {
+      "epoch": 2.28,
+      "learning_rate": 0.00016514657980456026,
+      "loss": 0.3255,
+      "step": 3550
+    },
+    {
+      "epoch": 2.29,
+      "learning_rate": 0.0001647557003257329,
+      "loss": 0.3367,
+      "step": 3560
+    },
+    {
+      "epoch": 2.3,
+      "learning_rate": 0.00016436482084690552,
+      "loss": 0.3403,
+      "step": 3570
+    },
+    {
+      "epoch": 2.3,
+      "learning_rate": 0.00016397394136807814,
+      "loss": 0.3294,
+      "step": 3580
+    },
+    {
+      "epoch": 2.31,
+      "learning_rate": 0.0001635830618892508,
+      "loss": 0.3334,
+      "step": 3590
+    },
+    {
+      "epoch": 2.32,
+      "learning_rate": 0.00016319218241042346,
+      "loss": 0.3207,
+      "step": 3600
+    },
+    {
+      "epoch": 2.32,
+      "eval_loss": 0.3624305725097656,
+      "eval_runtime": 264.5836,
+      "eval_samples_per_second": 7.559,
+      "eval_steps_per_second": 0.945,
+      "step": 3600
+    },
+    {
+      "epoch": 2.32,
+      "learning_rate": 0.00016280130293159608,
+      "loss": 0.3283,
+      "step": 3610
+    },
+    {
+      "epoch": 2.33,
+      "learning_rate": 0.0001624104234527687,
+      "loss": 0.3334,
+      "step": 3620
+    },
+    {
+      "epoch": 2.33,
+      "learning_rate": 0.00016201954397394134,
+      "loss": 0.3224,
+      "step": 3630
+    },
+    {
+      "epoch": 2.34,
+      "learning_rate": 0.000161628664495114,
+      "loss": 0.3229,
+      "step": 3640
+    },
+    {
+      "epoch": 2.35,
+      "learning_rate": 0.00016123778501628662,
+      "loss": 0.3377,
+      "step": 3650
+    },
+    {
+      "epoch": 2.35,
+      "learning_rate": 0.00016084690553745928,
+      "loss": 0.3247,
+      "step": 3660
+    },
+    {
+      "epoch": 2.36,
+      "learning_rate": 0.00016045602605863188,
+      "loss": 0.3328,
+      "step": 3670
+    },
+    {
+      "epoch": 2.37,
+      "learning_rate": 0.00016006514657980453,
+      "loss": 0.3402,
+      "step": 3680
+    },
+    {
+      "epoch": 2.37,
+      "learning_rate": 0.0001596742671009772,
+      "loss": 0.3233,
+      "step": 3690
+    },
+    {
+      "epoch": 2.38,
+      "learning_rate": 0.00015928338762214982,
+      "loss": 0.3281,
+      "step": 3700
+    },
+    {
+      "epoch": 2.39,
+      "learning_rate": 0.00015889250814332247,
+      "loss": 0.3325,
+      "step": 3710
+    },
+    {
+      "epoch": 2.39,
+      "learning_rate": 0.00015850162866449507,
+      "loss": 0.3335,
+      "step": 3720
+    },
+    {
+      "epoch": 2.4,
+      "learning_rate": 0.00015811074918566773,
+      "loss": 0.3266,
+      "step": 3730
+    },
+    {
+      "epoch": 2.41,
+      "learning_rate": 0.00015771986970684038,
+      "loss": 0.3536,
+      "step": 3740
+    },
+    {
+      "epoch": 2.41,
+      "learning_rate": 0.000157328990228013,
+      "loss": 0.335,
+      "step": 3750
+    },
+    {
+      "epoch": 2.42,
+      "learning_rate": 0.00015693811074918567,
+      "loss": 0.3369,
+      "step": 3760
+    },
+    {
+      "epoch": 2.42,
+      "learning_rate": 0.0001565472312703583,
+      "loss": 0.3227,
+      "step": 3770
+    },
+    {
+      "epoch": 2.43,
+      "learning_rate": 0.00015615635179153092,
+      "loss": 0.3207,
+      "step": 3780
+    },
+    {
+      "epoch": 2.44,
+      "learning_rate": 0.00015576547231270355,
+      "loss": 0.3217,
+      "step": 3790
+    },
+    {
+      "epoch": 2.44,
+      "learning_rate": 0.0001553745928338762,
+      "loss": 0.327,
+      "step": 3800
+    },
+    {
+      "epoch": 2.44,
+      "eval_loss": 0.3620395362377167,
+      "eval_runtime": 264.4714,
+      "eval_samples_per_second": 7.562,
+      "eval_steps_per_second": 0.945,
+      "step": 3800
+    },
+    {
+      "epoch": 2.45,
+      "learning_rate": 0.00015498371335504886,
+      "loss": 0.3285,
+      "step": 3810
+    },
+    {
+      "epoch": 2.46,
+      "learning_rate": 0.0001545928338762215,
+      "loss": 0.3254,
+      "step": 3820
+    },
+    {
+      "epoch": 2.46,
+      "learning_rate": 0.00015420195439739412,
+      "loss": 0.3364,
+      "step": 3830
+    },
+    {
+      "epoch": 2.47,
+      "learning_rate": 0.00015381107491856675,
+      "loss": 0.3326,
+      "step": 3840
+    },
+    {
+      "epoch": 2.48,
+      "learning_rate": 0.0001534201954397394,
+      "loss": 0.3226,
+      "step": 3850
+    },
+    {
+      "epoch": 2.48,
+      "learning_rate": 0.00015302931596091206,
+      "loss": 0.3243,
+      "step": 3860
+    },
+    {
+      "epoch": 2.49,
+      "learning_rate": 0.00015263843648208469,
+      "loss": 0.328,
+      "step": 3870
+    },
+    {
+      "epoch": 2.5,
+      "learning_rate": 0.00015224755700325731,
+      "loss": 0.3293,
+      "step": 3880
+    },
+    {
+      "epoch": 2.5,
+      "learning_rate": 0.00015185667752442994,
+      "loss": 0.3246,
+      "step": 3890
+    },
+    {
+      "epoch": 2.51,
+      "learning_rate": 0.0001514657980456026,
+      "loss": 0.3433,
+      "step": 3900
+    },
+    {
+      "epoch": 2.51,
+      "learning_rate": 0.00015107491856677523,
+      "loss": 0.3332,
+      "step": 3910
+    },
+    {
+      "epoch": 2.52,
+      "learning_rate": 0.00015068403908794788,
+      "loss": 0.3344,
+      "step": 3920
+    },
+    {
+      "epoch": 2.53,
+      "learning_rate": 0.00015029315960912048,
+      "loss": 0.3369,
+      "step": 3930
+    },
+    {
+      "epoch": 2.53,
+      "learning_rate": 0.00014990228013029314,
+      "loss": 0.3408,
+      "step": 3940
+    },
+    {
+      "epoch": 2.54,
+      "learning_rate": 0.0001495114006514658,
+      "loss": 0.3408,
+      "step": 3950
+    },
+    {
+      "epoch": 2.55,
+      "learning_rate": 0.00014912052117263842,
+      "loss": 0.3281,
+      "step": 3960
+    },
+    {
+      "epoch": 2.55,
+      "learning_rate": 0.00014872964169381105,
+      "loss": 0.3367,
+      "step": 3970
+    },
+    {
+      "epoch": 2.56,
+      "learning_rate": 0.0001483387622149837,
+      "loss": 0.3221,
+      "step": 3980
+    },
+    {
+      "epoch": 2.57,
+      "learning_rate": 0.00014794788273615633,
+      "loss": 0.3395,
+      "step": 3990
+    },
+    {
+      "epoch": 2.57,
+      "learning_rate": 0.000147557003257329,
+      "loss": 0.3374,
+      "step": 4000
+    },
+    {
+      "epoch": 2.57,
+      "eval_loss": 0.3611401319503784,
+      "eval_runtime": 264.4017,
+      "eval_samples_per_second": 7.564,
+      "eval_steps_per_second": 0.946,
+      "step": 4000
+    },
+    {
+      "epoch": 2.58,
+      "learning_rate": 0.00014716612377850162,
+      "loss": 0.3234,
+      "step": 4010
+    },
+    {
+      "epoch": 2.59,
+      "learning_rate": 0.00014677524429967424,
+      "loss": 0.354,
+      "step": 4020
+    },
+    {
+      "epoch": 2.59,
+      "learning_rate": 0.0001463843648208469,
+      "loss": 0.3396,
+      "step": 4030
+    },
+    {
+      "epoch": 2.6,
+      "learning_rate": 0.00014599348534201953,
+      "loss": 0.3326,
+      "step": 4040
+    },
+    {
+      "epoch": 2.6,
+      "learning_rate": 0.00014560260586319216,
+      "loss": 0.3201,
+      "step": 4050
+    },
+    {
+      "epoch": 2.61,
+      "learning_rate": 0.0001452117263843648,
+      "loss": 0.3281,
+      "step": 4060
+    },
+    {
+      "epoch": 2.62,
+      "learning_rate": 0.00014482084690553744,
+      "loss": 0.3313,
+      "step": 4070
+    },
+    {
+      "epoch": 2.62,
+      "learning_rate": 0.0001444299674267101,
+      "loss": 0.3466,
+      "step": 4080
+    },
+    {
+      "epoch": 2.63,
+      "learning_rate": 0.00014403908794788272,
+      "loss": 0.3277,
+      "step": 4090
+    },
+    {
+      "epoch": 2.64,
+      "learning_rate": 0.00014364820846905535,
+      "loss": 0.3381,
+      "step": 4100
+    },
+    {
+      "epoch": 2.64,
+      "learning_rate": 0.000143257328990228,
+      "loss": 0.3413,
+      "step": 4110
+    },
+    {
+      "epoch": 2.65,
+      "learning_rate": 0.00014286644951140063,
+      "loss": 0.3347,
+      "step": 4120
+    },
+    {
+      "epoch": 2.66,
+      "learning_rate": 0.0001424755700325733,
+      "loss": 0.3338,
+      "step": 4130
+    },
+    {
+      "epoch": 2.66,
+      "learning_rate": 0.00014208469055374592,
+      "loss": 0.3339,
+      "step": 4140
+    },
+    {
+      "epoch": 2.67,
+      "learning_rate": 0.00014169381107491854,
+      "loss": 0.3406,
+      "step": 4150
+    },
+    {
+      "epoch": 2.68,
+      "learning_rate": 0.0001413029315960912,
+      "loss": 0.3348,
+      "step": 4160
+    },
+    {
+      "epoch": 2.68,
+      "learning_rate": 0.00014091205211726383,
+      "loss": 0.3234,
+      "step": 4170
+    },
+    {
+      "epoch": 2.69,
+      "learning_rate": 0.00014052117263843646,
+      "loss": 0.3336,
+      "step": 4180
+    },
+    {
+      "epoch": 2.69,
+      "learning_rate": 0.0001401302931596091,
+      "loss": 0.3381,
+      "step": 4190
+    },
+    {
+      "epoch": 2.7,
+      "learning_rate": 0.00013973941368078174,
+      "loss": 0.3299,
+      "step": 4200
+    },
+    {
+      "epoch": 2.7,
+      "eval_loss": 0.3604828417301178,
+      "eval_runtime": 264.3421,
+      "eval_samples_per_second": 7.566,
+      "eval_steps_per_second": 0.946,
+      "step": 4200
+    },
+    {
+      "epoch": 2.71,
+      "learning_rate": 0.0001393485342019544,
+      "loss": 0.3258,
+      "step": 4210
+    },
+    {
+      "epoch": 2.71,
+      "learning_rate": 0.00013895765472312702,
+      "loss": 0.3415,
+      "step": 4220
+    },
+    {
+      "epoch": 2.72,
+      "learning_rate": 0.00013856677524429965,
+      "loss": 0.3155,
+      "step": 4230
+    },
+    {
+      "epoch": 2.73,
+      "learning_rate": 0.0001381758957654723,
+      "loss": 0.3467,
+      "step": 4240
+    },
+    {
+      "epoch": 2.73,
+      "learning_rate": 0.00013778501628664493,
+      "loss": 0.333,
+      "step": 4250
+    },
+    {
+      "epoch": 2.74,
+      "learning_rate": 0.0001373941368078176,
+      "loss": 0.3287,
+      "step": 4260
+    },
+    {
+      "epoch": 2.75,
+      "learning_rate": 0.00013700325732899022,
+      "loss": 0.3294,
+      "step": 4270
+    },
+    {
+      "epoch": 2.75,
+      "learning_rate": 0.00013661237785016285,
+      "loss": 0.3432,
+      "step": 4280
+    },
+    {
+      "epoch": 2.76,
+      "learning_rate": 0.0001362214983713355,
+      "loss": 0.3307,
+      "step": 4290
+    },
+    {
+      "epoch": 2.77,
+      "learning_rate": 0.00013583061889250813,
+      "loss": 0.3265,
+      "step": 4300
+    },
+    {
+      "epoch": 2.77,
+      "learning_rate": 0.00013543973941368076,
+      "loss": 0.3389,
+      "step": 4310
+    },
+    {
+      "epoch": 2.78,
+      "learning_rate": 0.0001350488599348534,
+      "loss": 0.347,
+      "step": 4320
+    },
+    {
+      "epoch": 2.78,
+      "learning_rate": 0.00013465798045602604,
+      "loss": 0.3372,
+      "step": 4330
+    },
+    {
+      "epoch": 2.79,
+      "learning_rate": 0.0001342671009771987,
+      "loss": 0.3257,
+      "step": 4340
+    },
+    {
+      "epoch": 2.8,
+      "learning_rate": 0.00013387622149837132,
+      "loss": 0.3331,
+      "step": 4350
+    },
+    {
+      "epoch": 2.8,
+      "learning_rate": 0.00013348534201954395,
+      "loss": 0.3385,
+      "step": 4360
+    },
+    {
+      "epoch": 2.81,
+      "learning_rate": 0.0001330944625407166,
+      "loss": 0.3314,
+      "step": 4370
+    },
+    {
+      "epoch": 2.82,
+      "learning_rate": 0.00013270358306188924,
+      "loss": 0.339,
+      "step": 4380
+    },
+    {
+      "epoch": 2.82,
+      "learning_rate": 0.0001323127035830619,
+      "loss": 0.3294,
+      "step": 4390
+    },
+    {
+      "epoch": 2.83,
+      "learning_rate": 0.00013192182410423452,
+      "loss": 0.3383,
+      "step": 4400
+    },
+    {
+      "epoch": 2.83,
+      "eval_loss": 0.35994288325309753,
+      "eval_runtime": 264.3191,
+      "eval_samples_per_second": 7.567,
+      "eval_steps_per_second": 0.946,
+      "step": 4400
+    },
+    {
+      "epoch": 2.84,
+      "learning_rate": 0.00013153094462540715,
+      "loss": 0.3207,
+      "step": 4410
+    },
+    {
+      "epoch": 2.84,
+      "learning_rate": 0.0001311400651465798,
+      "loss": 0.3348,
+      "step": 4420
+    },
+    {
+      "epoch": 2.85,
+      "learning_rate": 0.00013074918566775243,
+      "loss": 0.3238,
+      "step": 4430
+    },
+    {
+      "epoch": 2.86,
+      "learning_rate": 0.00013035830618892506,
+      "loss": 0.3231,
+      "step": 4440
+    },
+    {
+      "epoch": 2.86,
+      "learning_rate": 0.00012996742671009771,
+      "loss": 0.3241,
+      "step": 4450
+    },
+    {
+      "epoch": 2.87,
+      "learning_rate": 0.00012957654723127034,
+      "loss": 0.3287,
+      "step": 4460
+    },
+    {
+      "epoch": 2.87,
+      "learning_rate": 0.000129185667752443,
+      "loss": 0.3308,
+      "step": 4470
+    },
+    {
+      "epoch": 2.88,
+      "learning_rate": 0.00012879478827361563,
+      "loss": 0.3432,
+      "step": 4480
+    },
+    {
+      "epoch": 2.89,
+      "learning_rate": 0.00012840390879478825,
+      "loss": 0.3435,
+      "step": 4490
+    },
+    {
+      "epoch": 2.89,
+      "learning_rate": 0.0001280130293159609,
+      "loss": 0.3158,
+      "step": 4500
+    },
+    {
+      "epoch": 2.9,
+      "learning_rate": 0.00012762214983713354,
+      "loss": 0.3453,
+      "step": 4510
+    },
+    {
+      "epoch": 2.91,
+      "learning_rate": 0.0001272312703583062,
+      "loss": 0.3317,
+      "step": 4520
+    },
+    {
+      "epoch": 2.91,
+      "learning_rate": 0.00012684039087947882,
+      "loss": 0.3555,
+      "step": 4530
+    },
+    {
+      "epoch": 2.92,
+      "learning_rate": 0.00012644951140065145,
+      "loss": 0.3442,
+      "step": 4540
+    },
+    {
+      "epoch": 2.93,
+      "learning_rate": 0.0001260586319218241,
+      "loss": 0.3264,
+      "step": 4550
+    },
+    {
+      "epoch": 2.93,
+      "learning_rate": 0.00012566775244299673,
+      "loss": 0.3239,
+      "step": 4560
+    },
+    {
+      "epoch": 2.94,
+      "learning_rate": 0.00012527687296416936,
+      "loss": 0.3239,
+      "step": 4570
+    },
+    {
+      "epoch": 2.95,
+      "learning_rate": 0.000124885993485342,
+      "loss": 0.3449,
+      "step": 4580
+    },
+    {
+      "epoch": 2.95,
+      "learning_rate": 0.00012449511400651464,
+      "loss": 0.3277,
+      "step": 4590
+    },
+    {
+      "epoch": 2.96,
+      "learning_rate": 0.0001241042345276873,
+      "loss": 0.3396,
+      "step": 4600
+    },
+    {
+      "epoch": 2.96,
+      "eval_loss": 0.35899612307548523,
+      "eval_runtime": 264.3094,
+      "eval_samples_per_second": 7.567,
+      "eval_steps_per_second": 0.946,
+      "step": 4600
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 7775,
+  "num_train_epochs": 5,
+  "save_steps": 200,
+  "total_flos": 3.087947012827644e+18,
+  "trial_name": null,
+  "trial_params": null
+}

with_input/decomp_code/best_model/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:93f9fbac1684a8649409617438605549f49d48e0a19fe17193c7b46907449261
+size 4027