Upload folder using huggingface_hub

checkpoint-1400/adapter_config.json

@@ -10,20 +10,18 @@
   "layers_pattern": null,
   "layers_to_transform": null,
   "loftq_config": {},
-  "lora_alpha": 16,
+  "lora_alpha": 32,
   "lora_dropout": 0.1,
   "megatron_config": null,
   "megatron_core": "megatron.core",
   "modules_to_save": null,
   "peft_type": "LORA",
-  "r": 64,
+  "r": 8,
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
     "v_proj",
-    "o_proj",
-    "q_proj",
-    "k_proj"
+    "q_proj"
   ],
   "task_type": "CAUSAL_LM",
   "use_dora": false,

checkpoint-1400/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c9401d1e629b1ab3d58f131e884be617a6d85bda9188213b3fe9a349e5b3de0c
-size 54543112
+oid sha256:1be083b1e7d0a2ca73501e3a1fbce6b84a850044a233872d5b01cdcf0a65c58f
+size 3416264

checkpoint-1400/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4f025e413edc76837904c1e5cfc24e7d486f79382a2b2fdbfd642c2343b983d4
-size 109159930
+oid sha256:642c174b47359c6527c32c721dc0bc13b6b0b98189bbc690582671c72af6696b
+size 6869818

checkpoint-1400/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6d5b74bc90e579e47da3877ca4510f5d87b4db22e3422b487a0da42e750bff90
+oid sha256:6e654950e8ee96fc7cda636036720d303346a436a535ba00c826eceac02a302f
 size 14244

checkpoint-1400/trainer_state.json

@@ -10,990 +10,990 @@
   "log_history": [
     {
       "epoch": 0.007142857142857143,
-      "grad_norm": 0.8348605036735535,
+      "grad_norm": 3.0706775188446045,
       "learning_rate": 9.92857142857143e-05,
-      "loss": 1.2652,
+      "loss": 1.2883,
       "step": 10
     },
     {
       "epoch": 0.014285714285714285,
-      "grad_norm": 0.41527849435806274,
+      "grad_norm": 2.5454561710357666,
       "learning_rate": 9.857142857142858e-05,
-      "loss": 1.1228,
+      "loss": 1.1211,
       "step": 20
     },
     {
       "epoch": 0.02142857142857143,
-      "grad_norm": 0.7952600717544556,
+      "grad_norm": 2.717937469482422,
       "learning_rate": 9.785714285714286e-05,
-      "loss": 0.9764,
+      "loss": 0.9927,
       "step": 30
     },
     {
       "epoch": 0.02857142857142857,
-      "grad_norm": 0.2524937689304352,
+      "grad_norm": 1.2683441638946533,
       "learning_rate": 9.714285714285715e-05,
-      "loss": 0.887,
+      "loss": 0.9406,
       "step": 40
     },
     {
       "epoch": 0.03571428571428571,
-      "grad_norm": 0.7529658675193787,
+      "grad_norm": 7.902857303619385,
       "learning_rate": 9.642857142857143e-05,
-      "loss": 0.6871,
+      "loss": 0.7558,
       "step": 50
     },
     {
       "epoch": 0.04285714285714286,
-      "grad_norm": 0.289154052734375,
+      "grad_norm": 2.552997350692749,
       "learning_rate": 9.571428571428573e-05,
-      "loss": 0.5856,
+      "loss": 0.6445,
       "step": 60
     },
     {
       "epoch": 0.05,
-      "grad_norm": 0.6982300281524658,
+      "grad_norm": 3.5295348167419434,
       "learning_rate": 9.5e-05,
-      "loss": 0.6057,
+      "loss": 0.6534,
       "step": 70
     },
     {
       "epoch": 0.05714285714285714,
-      "grad_norm": 0.36802777647972107,
+      "grad_norm": 1.8732738494873047,
       "learning_rate": 9.428571428571429e-05,
-      "loss": 0.5035,
+      "loss": 0.537,
       "step": 80
     },
     {
       "epoch": 0.06428571428571428,
-      "grad_norm": 0.7616478800773621,
+      "grad_norm": 4.299726963043213,
       "learning_rate": 9.357142857142858e-05,
-      "loss": 0.5143,
+      "loss": 0.5508,
       "step": 90
     },
     {
       "epoch": 0.07142857142857142,
-      "grad_norm": 0.36089685559272766,
+      "grad_norm": 2.1415882110595703,
       "learning_rate": 9.285714285714286e-05,
-      "loss": 0.5063,
+      "loss": 0.5438,
       "step": 100
     },
     {
       "epoch": 0.07857142857142857,
-      "grad_norm": 0.31661227345466614,
+      "grad_norm": 3.7711620330810547,
       "learning_rate": 9.214285714285714e-05,
-      "loss": 0.5535,
+      "loss": 0.5906,
       "step": 110
     },
     {
       "epoch": 0.08571428571428572,
-      "grad_norm": 0.19921253621578217,
+      "grad_norm": 1.4387797117233276,
       "learning_rate": 9.142857142857143e-05,
-      "loss": 0.5267,
+      "loss": 0.541,
       "step": 120
     },
     {
       "epoch": 0.09285714285714286,
-      "grad_norm": 0.4584966003894806,
+      "grad_norm": 4.010119915008545,
       "learning_rate": 9.071428571428571e-05,
-      "loss": 0.5315,
+      "loss": 0.56,
       "step": 130
     },
     {
       "epoch": 0.1,
-      "grad_norm": 0.4068152606487274,
+      "grad_norm": 4.021274089813232,
       "learning_rate": 9e-05,
-      "loss": 0.5072,
+      "loss": 0.5323,
       "step": 140
     },
     {
       "epoch": 0.10714285714285714,
-      "grad_norm": 0.40502503514289856,
+      "grad_norm": 2.138714551925659,
       "learning_rate": 8.92857142857143e-05,
-      "loss": 0.4591,
+      "loss": 0.4784,
       "step": 150
     },
     {
       "epoch": 0.11428571428571428,
-      "grad_norm": 0.22944864630699158,
+      "grad_norm": 1.371328353881836,
       "learning_rate": 8.857142857142857e-05,
-      "loss": 0.4721,
+      "loss": 0.4861,
       "step": 160
     },
     {
       "epoch": 0.12142857142857143,
-      "grad_norm": 0.39912867546081543,
+      "grad_norm": 3.1777868270874023,
       "learning_rate": 8.785714285714286e-05,
-      "loss": 0.4958,
+      "loss": 0.504,
       "step": 170
     },
     {
       "epoch": 0.12857142857142856,
-      "grad_norm": 0.4304330050945282,
+      "grad_norm": 2.0382461547851562,
       "learning_rate": 8.714285714285715e-05,
-      "loss": 0.4957,
+      "loss": 0.496,
       "step": 180
     },
     {
       "epoch": 0.1357142857142857,
-      "grad_norm": 0.6040734648704529,
+      "grad_norm": 2.8927695751190186,
       "learning_rate": 8.642857142857143e-05,
-      "loss": 0.5536,
+      "loss": 0.5581,
       "step": 190
     },
     {
       "epoch": 0.14285714285714285,
-      "grad_norm": 0.7617625594139099,
+      "grad_norm": 4.461900234222412,
       "learning_rate": 8.571428571428571e-05,
-      "loss": 0.453,
+      "loss": 0.4643,
       "step": 200
     },
     {
       "epoch": 0.15,
-      "grad_norm": 0.30004259943962097,
+      "grad_norm": 2.028140068054199,
       "learning_rate": 8.5e-05,
-      "loss": 0.4292,
+      "loss": 0.4384,
       "step": 210
     },
     {
       "epoch": 0.15714285714285714,
-      "grad_norm": 0.6278651356697083,
+      "grad_norm": 2.4652626514434814,
       "learning_rate": 8.428571428571429e-05,
-      "loss": 0.4667,
+      "loss": 0.4705,
       "step": 220
     },
     {
       "epoch": 0.16428571428571428,
-      "grad_norm": 0.4061867892742157,
+      "grad_norm": 2.360128879547119,
       "learning_rate": 8.357142857142858e-05,
-      "loss": 0.4267,
+      "loss": 0.4238,
       "step": 230
     },
     {
       "epoch": 0.17142857142857143,
-      "grad_norm": 0.2937834858894348,
+      "grad_norm": 1.5631648302078247,
       "learning_rate": 8.285714285714287e-05,
-      "loss": 0.4398,
+      "loss": 0.4364,
       "step": 240
     },
     {
       "epoch": 0.17857142857142858,
-      "grad_norm": 0.41685324907302856,
+      "grad_norm": 1.9402961730957031,
       "learning_rate": 8.214285714285714e-05,
-      "loss": 0.4409,
+      "loss": 0.4472,
       "step": 250
     },
     {
       "epoch": 0.18571428571428572,
-      "grad_norm": 0.5377163887023926,
+      "grad_norm": 2.9574289321899414,
       "learning_rate": 8.142857142857143e-05,
-      "loss": 0.4335,
+      "loss": 0.4319,
       "step": 260
     },
     {
       "epoch": 0.19285714285714287,
-      "grad_norm": 0.2825513184070587,
+      "grad_norm": 1.7986479997634888,
       "learning_rate": 8.071428571428573e-05,
-      "loss": 0.4841,
+      "loss": 0.4873,
       "step": 270
     },
     {
       "epoch": 0.2,
-      "grad_norm": 0.6520470380783081,
+      "grad_norm": 4.539698123931885,
       "learning_rate": 8e-05,
-      "loss": 0.427,
+      "loss": 0.4222,
       "step": 280
     },
     {
       "epoch": 0.20714285714285716,
-      "grad_norm": 0.6439694762229919,
+      "grad_norm": 4.539310932159424,
       "learning_rate": 7.928571428571429e-05,
-      "loss": 0.4373,
+      "loss": 0.444,
       "step": 290
     },
     {
       "epoch": 0.21428571428571427,
-      "grad_norm": 0.7449022531509399,
+      "grad_norm": 4.211949348449707,
       "learning_rate": 7.857142857142858e-05,
-      "loss": 0.4622,
+      "loss": 0.4713,
       "step": 300
     },
     {
       "epoch": 0.22142857142857142,
-      "grad_norm": 0.702427864074707,
+      "grad_norm": 3.4612796306610107,
       "learning_rate": 7.785714285714286e-05,
-      "loss": 0.4139,
+      "loss": 0.421,
       "step": 310
     },
     {
       "epoch": 0.22857142857142856,
-      "grad_norm": 0.4422167241573334,
+      "grad_norm": 2.1370456218719482,
       "learning_rate": 7.714285714285715e-05,
-      "loss": 0.5105,
+      "loss": 0.5117,
       "step": 320
     },
     {
       "epoch": 0.2357142857142857,
-      "grad_norm": 0.3643128573894501,
+      "grad_norm": 2.022341012954712,
       "learning_rate": 7.642857142857143e-05,
-      "loss": 0.4018,
+      "loss": 0.3968,
       "step": 330
     },
     {
       "epoch": 0.24285714285714285,
-      "grad_norm": 0.3324808180332184,
+      "grad_norm": 1.5263112783432007,
       "learning_rate": 7.571428571428571e-05,
-      "loss": 0.3991,
+      "loss": 0.4032,
       "step": 340
     },
     {
       "epoch": 0.25,
-      "grad_norm": 0.5424559116363525,
+      "grad_norm": 3.152087450027466,
       "learning_rate": 7.500000000000001e-05,
-      "loss": 0.3948,
+      "loss": 0.4024,
       "step": 350
     },
     {
       "epoch": 0.2571428571428571,
-      "grad_norm": 0.7667222619056702,
+      "grad_norm": 3.5288479328155518,
       "learning_rate": 7.428571428571429e-05,
-      "loss": 0.4075,
+      "loss": 0.4079,
       "step": 360
     },
     {
       "epoch": 0.2642857142857143,
-      "grad_norm": 0.3680099844932556,
+      "grad_norm": 1.922340989112854,
       "learning_rate": 7.357142857142858e-05,
-      "loss": 0.4721,
+      "loss": 0.4839,
       "step": 370
     },
     {
       "epoch": 0.2714285714285714,
-      "grad_norm": 0.4167829155921936,
+      "grad_norm": 2.6698594093322754,
       "learning_rate": 7.285714285714286e-05,
-      "loss": 0.3657,
+      "loss": 0.3772,
       "step": 380
     },
     {
       "epoch": 0.2785714285714286,
-      "grad_norm": 0.6765364408493042,
+      "grad_norm": 2.9632909297943115,
       "learning_rate": 7.214285714285714e-05,
-      "loss": 0.4158,
+      "loss": 0.4296,
       "step": 390
     },
     {
       "epoch": 0.2857142857142857,
-      "grad_norm": 0.5632287859916687,
+      "grad_norm": 2.624523639678955,
       "learning_rate": 7.142857142857143e-05,
-      "loss": 0.3808,
+      "loss": 0.3931,
       "step": 400
     },
     {
       "epoch": 0.29285714285714287,
-      "grad_norm": 0.6470784544944763,
+      "grad_norm": 3.6488404273986816,
       "learning_rate": 7.071428571428573e-05,
-      "loss": 0.3818,
+      "loss": 0.3986,
       "step": 410
     },
     {
       "epoch": 0.3,
-      "grad_norm": 0.6362716555595398,
+      "grad_norm": 3.6739540100097656,
       "learning_rate": 7e-05,
-      "loss": 0.382,
+      "loss": 0.3979,
       "step": 420
     },
     {
       "epoch": 0.30714285714285716,
-      "grad_norm": 0.7216530442237854,
+      "grad_norm": 2.866800546646118,
       "learning_rate": 6.928571428571429e-05,
-      "loss": 0.4106,
+      "loss": 0.4203,
       "step": 430
     },
     {
       "epoch": 0.3142857142857143,
-      "grad_norm": 0.6893362402915955,
+      "grad_norm": 3.4130825996398926,
       "learning_rate": 6.857142857142858e-05,
-      "loss": 0.4753,
+      "loss": 0.498,
       "step": 440
     },
     {
       "epoch": 0.32142857142857145,
-      "grad_norm": 0.3167782127857208,
+      "grad_norm": 1.6507424116134644,
       "learning_rate": 6.785714285714286e-05,
-      "loss": 0.3535,
+      "loss": 0.3672,
       "step": 450
     },
     {
       "epoch": 0.32857142857142857,
-      "grad_norm": 0.5213643908500671,
+      "grad_norm": 2.9661030769348145,
       "learning_rate": 6.714285714285714e-05,
-      "loss": 0.355,
+      "loss": 0.372,
       "step": 460
     },
     {
       "epoch": 0.3357142857142857,
-      "grad_norm": 0.44351714849472046,
+      "grad_norm": 1.8931576013565063,
       "learning_rate": 6.642857142857143e-05,
-      "loss": 0.3341,
+      "loss": 0.341,
       "step": 470
     },
     {
       "epoch": 0.34285714285714286,
-      "grad_norm": 0.45469874143600464,
+      "grad_norm": 2.128786563873291,
       "learning_rate": 6.571428571428571e-05,
-      "loss": 0.318,
+      "loss": 0.3235,
       "step": 480
     },
     {
       "epoch": 0.35,
-      "grad_norm": 0.6794070601463318,
+      "grad_norm": 3.9210121631622314,
       "learning_rate": 6.500000000000001e-05,
-      "loss": 0.4363,
+      "loss": 0.4568,
       "step": 490
     },
     {
       "epoch": 0.35714285714285715,
-      "grad_norm": 0.35572198033332825,
+      "grad_norm": 2.2050564289093018,
       "learning_rate": 6.428571428571429e-05,
-      "loss": 0.337,
+      "loss": 0.3468,
       "step": 500
     },
     {
       "epoch": 0.36428571428571427,
-      "grad_norm": 0.2703068256378174,
+      "grad_norm": 1.4351139068603516,
       "learning_rate": 6.357142857142857e-05,
-      "loss": 0.3572,
+      "loss": 0.3745,
       "step": 510
     },
     {
       "epoch": 0.37142857142857144,
-      "grad_norm": 0.37365567684173584,
+      "grad_norm": 1.60938560962677,
       "learning_rate": 6.285714285714286e-05,
-      "loss": 0.3851,
+      "loss": 0.3944,
       "step": 520
     },
     {
       "epoch": 0.37857142857142856,
-      "grad_norm": 0.623077392578125,
+      "grad_norm": 3.240180730819702,
       "learning_rate": 6.214285714285714e-05,
-      "loss": 0.352,
+      "loss": 0.3701,
       "step": 530
     },
     {
       "epoch": 0.38571428571428573,
-      "grad_norm": 0.721612274646759,
+      "grad_norm": 2.8629231452941895,
       "learning_rate": 6.142857142857143e-05,
-      "loss": 0.3283,
+      "loss": 0.3378,
       "step": 540
     },
     {
       "epoch": 0.39285714285714285,
-      "grad_norm": 0.45066478848457336,
+      "grad_norm": 2.2353878021240234,
       "learning_rate": 6.0714285714285715e-05,
-      "loss": 0.3644,
+      "loss": 0.3783,
       "step": 550
     },
     {
       "epoch": 0.4,
-      "grad_norm": 0.5320075154304504,
+      "grad_norm": 5.000092029571533,
       "learning_rate": 6e-05,
-      "loss": 0.3458,
+      "loss": 0.3664,
       "step": 560
     },
     {
       "epoch": 0.40714285714285714,
-      "grad_norm": 0.669303297996521,
+      "grad_norm": 1.9683377742767334,
       "learning_rate": 5.928571428571429e-05,
-      "loss": 0.4307,
+      "loss": 0.4366,
       "step": 570
     },
     {
       "epoch": 0.4142857142857143,
-      "grad_norm": 0.4207151234149933,
+      "grad_norm": 1.6545411348342896,
       "learning_rate": 5.8571428571428575e-05,
-      "loss": 0.3836,
+      "loss": 0.4047,
       "step": 580
     },
     {
       "epoch": 0.42142857142857143,
-      "grad_norm": 0.992280125617981,
+      "grad_norm": 6.393994331359863,
       "learning_rate": 5.785714285714287e-05,
-      "loss": 0.3339,
+      "loss": 0.3513,
       "step": 590
     },
     {
       "epoch": 0.42857142857142855,
-      "grad_norm": 0.9569453001022339,
+      "grad_norm": 2.786479949951172,
       "learning_rate": 5.714285714285714e-05,
-      "loss": 0.3794,
+      "loss": 0.3973,
       "step": 600
     },
     {
       "epoch": 0.4357142857142857,
-      "grad_norm": 0.48947492241859436,
+      "grad_norm": 2.1006624698638916,
       "learning_rate": 5.642857142857143e-05,
-      "loss": 0.4,
+      "loss": 0.413,
       "step": 610
     },
     {
       "epoch": 0.44285714285714284,
-      "grad_norm": 0.5944671034812927,
+      "grad_norm": 2.376877784729004,
       "learning_rate": 5.571428571428572e-05,
-      "loss": 0.4114,
+      "loss": 0.427,
       "step": 620
     },
     {
       "epoch": 0.45,
-      "grad_norm": 1.0119761228561401,
+      "grad_norm": 3.311000347137451,
       "learning_rate": 5.500000000000001e-05,
-      "loss": 0.3887,
+      "loss": 0.4043,
       "step": 630
     },
     {
       "epoch": 0.45714285714285713,
-      "grad_norm": 1.074351191520691,
+      "grad_norm": 5.115108489990234,
       "learning_rate": 5.428571428571428e-05,
-      "loss": 0.4314,
+      "loss": 0.4501,
       "step": 640
     },
     {
       "epoch": 0.4642857142857143,
-      "grad_norm": 0.36413517594337463,
+      "grad_norm": 1.7104153633117676,
       "learning_rate": 5.3571428571428575e-05,
-      "loss": 0.3711,
+      "loss": 0.3966,
       "step": 650
     },
     {
       "epoch": 0.4714285714285714,
-      "grad_norm": 0.43678954243659973,
+      "grad_norm": 1.9817299842834473,
       "learning_rate": 5.285714285714286e-05,
-      "loss": 0.3162,
+      "loss": 0.3308,
       "step": 660
     },
     {
       "epoch": 0.4785714285714286,
-      "grad_norm": 0.5720781683921814,
+      "grad_norm": 2.1275410652160645,
       "learning_rate": 5.214285714285715e-05,
-      "loss": 0.303,
+      "loss": 0.3219,
       "step": 670
     },
     {
       "epoch": 0.4857142857142857,
-      "grad_norm": 0.59657883644104,
+      "grad_norm": 2.4896678924560547,
       "learning_rate": 5.142857142857143e-05,
-      "loss": 0.2921,
+      "loss": 0.3084,
       "step": 680
     },
     {
       "epoch": 0.4928571428571429,
-      "grad_norm": 0.4800083637237549,
+      "grad_norm": 2.4593498706817627,
       "learning_rate": 5.0714285714285716e-05,
-      "loss": 0.3375,
+      "loss": 0.3695,
       "step": 690
     },
     {
       "epoch": 0.5,
-      "grad_norm": 0.5119388699531555,
+      "grad_norm": 2.4440925121307373,
       "learning_rate": 5e-05,
-      "loss": 0.337,
+      "loss": 0.3584,
       "step": 700
     },
     {
       "epoch": 0.5071428571428571,
-      "grad_norm": 0.43967753648757935,
+      "grad_norm": 1.6439753770828247,
       "learning_rate": 4.928571428571429e-05,
-      "loss": 0.4005,
+      "loss": 0.4272,
       "step": 710
     },
     {
       "epoch": 0.5142857142857142,
-      "grad_norm": 0.4737344980239868,
+      "grad_norm": 2.897373676300049,
       "learning_rate": 4.8571428571428576e-05,
-      "loss": 0.3365,
+      "loss": 0.3496,
       "step": 720
     },
     {
       "epoch": 0.5214285714285715,
-      "grad_norm": 0.6178275346755981,
+      "grad_norm": 2.620026111602783,
       "learning_rate": 4.785714285714286e-05,
-      "loss": 0.3612,
+      "loss": 0.3745,
       "step": 730
     },
     {
       "epoch": 0.5285714285714286,
-      "grad_norm": 0.45657363533973694,
+      "grad_norm": 2.081660032272339,
       "learning_rate": 4.714285714285714e-05,
-      "loss": 0.3474,
+      "loss": 0.3608,
       "step": 740
     },
     {
       "epoch": 0.5357142857142857,
-      "grad_norm": 0.3173658847808838,
+      "grad_norm": 1.8209961652755737,
       "learning_rate": 4.642857142857143e-05,
-      "loss": 0.3295,
+      "loss": 0.3429,
       "step": 750
     },
     {
       "epoch": 0.5428571428571428,
-      "grad_norm": 0.5634258389472961,
+      "grad_norm": 1.898141622543335,
       "learning_rate": 4.5714285714285716e-05,
-      "loss": 0.347,
+      "loss": 0.3671,
       "step": 760
     },
     {
       "epoch": 0.55,
-      "grad_norm": 0.48927634954452515,
+      "grad_norm": 2.252241611480713,
       "learning_rate": 4.5e-05,
-      "loss": 0.3665,
+      "loss": 0.3843,
       "step": 770
     },
     {
       "epoch": 0.5571428571428572,
-      "grad_norm": 0.4992411732673645,
+      "grad_norm": 2.201845407485962,
       "learning_rate": 4.428571428571428e-05,
-      "loss": 0.3229,
+      "loss": 0.3379,
       "step": 780
     },
     {
       "epoch": 0.5642857142857143,
-      "grad_norm": 0.2783334255218506,
+      "grad_norm": 1.9457173347473145,
       "learning_rate": 4.3571428571428576e-05,
-      "loss": 0.3176,
+      "loss": 0.3366,
       "step": 790
     },
     {
       "epoch": 0.5714285714285714,
-      "grad_norm": 0.7770132422447205,
+      "grad_norm": 3.7929930686950684,
       "learning_rate": 4.2857142857142856e-05,
-      "loss": 0.292,
+      "loss": 0.3091,
       "step": 800
     },
     {
       "epoch": 0.5785714285714286,
-      "grad_norm": 0.3741398751735687,
+      "grad_norm": 1.8005012273788452,
       "learning_rate": 4.214285714285714e-05,
-      "loss": 0.2749,
+      "loss": 0.2944,
       "step": 810
     },
     {
       "epoch": 0.5857142857142857,
-      "grad_norm": 1.1872639656066895,
+      "grad_norm": 6.003371715545654,
       "learning_rate": 4.1428571428571437e-05,
-      "loss": 0.3475,
+      "loss": 0.3722,
       "step": 820
     },
     {
       "epoch": 0.5928571428571429,
-      "grad_norm": 0.5437417030334473,
+      "grad_norm": 2.1478536128997803,
       "learning_rate": 4.0714285714285717e-05,
-      "loss": 0.2855,
+      "loss": 0.3081,
       "step": 830
     },
     {
       "epoch": 0.6,
-      "grad_norm": 0.6971333622932434,
+      "grad_norm": 2.839242935180664,
       "learning_rate": 4e-05,
-      "loss": 0.3026,
+      "loss": 0.317,
       "step": 840
     },
     {
       "epoch": 0.6071428571428571,
-      "grad_norm": 0.4371561110019684,
+      "grad_norm": 1.891158938407898,
       "learning_rate": 3.928571428571429e-05,
-      "loss": 0.2841,
+      "loss": 0.2995,
       "step": 850
     },
     {
       "epoch": 0.6142857142857143,
-      "grad_norm": 0.7760091423988342,
+      "grad_norm": 5.433562278747559,
       "learning_rate": 3.857142857142858e-05,
-      "loss": 0.347,
+      "loss": 0.3637,
       "step": 860
     },
     {
       "epoch": 0.6214285714285714,
-      "grad_norm": 0.29647722840309143,
+      "grad_norm": 1.653617024421692,
       "learning_rate": 3.785714285714286e-05,
-      "loss": 0.3093,
+      "loss": 0.3285,
       "step": 870
     },
     {
       "epoch": 0.6285714285714286,
-      "grad_norm": 0.5774063467979431,
+      "grad_norm": 2.9657585620880127,
       "learning_rate": 3.7142857142857143e-05,
-      "loss": 0.355,
+      "loss": 0.3761,
       "step": 880
     },
     {
       "epoch": 0.6357142857142857,
-      "grad_norm": 0.42540696263313293,
+      "grad_norm": 1.5580040216445923,
       "learning_rate": 3.642857142857143e-05,
-      "loss": 0.4248,
+      "loss": 0.4427,
       "step": 890
     },
     {
       "epoch": 0.6428571428571429,
-      "grad_norm": 0.4201267659664154,
+      "grad_norm": 2.411190986633301,
       "learning_rate": 3.571428571428572e-05,
-      "loss": 0.3186,
+      "loss": 0.3322,
       "step": 900
     },
     {
       "epoch": 0.65,
-      "grad_norm": 0.6699719429016113,
+      "grad_norm": 4.375690937042236,
       "learning_rate": 3.5e-05,
-      "loss": 0.3345,
+      "loss": 0.3546,
       "step": 910
     },
     {
       "epoch": 0.6571428571428571,
-      "grad_norm": 0.9379881024360657,
+      "grad_norm": 3.571958541870117,
       "learning_rate": 3.428571428571429e-05,
-      "loss": 0.3523,
+      "loss": 0.3742,
       "step": 920
     },
     {
       "epoch": 0.6642857142857143,
-      "grad_norm": 0.4937261939048767,
+      "grad_norm": 3.132997512817383,
       "learning_rate": 3.357142857142857e-05,
-      "loss": 0.3424,
+      "loss": 0.3667,
       "step": 930
     },
     {
       "epoch": 0.6714285714285714,
-      "grad_norm": 0.4084944427013397,
+      "grad_norm": 2.296008586883545,
       "learning_rate": 3.285714285714286e-05,
-      "loss": 0.2488,
+      "loss": 0.2686,
       "step": 940
     },
     {
       "epoch": 0.6785714285714286,
-      "grad_norm": 0.3513486981391907,
+      "grad_norm": 1.7339441776275635,
       "learning_rate": 3.2142857142857144e-05,
-      "loss": 0.308,
+      "loss": 0.328,
       "step": 950
     },
     {
       "epoch": 0.6857142857142857,
-      "grad_norm": 0.6102254986763,
+      "grad_norm": 2.5679969787597656,
       "learning_rate": 3.142857142857143e-05,
-      "loss": 0.326,
+      "loss": 0.3437,
       "step": 960
     },
     {
       "epoch": 0.6928571428571428,
-      "grad_norm": 1.241847038269043,
+      "grad_norm": 6.115095615386963,
       "learning_rate": 3.071428571428572e-05,
-      "loss": 0.3028,
+      "loss": 0.3264,
       "step": 970
     },
     {
       "epoch": 0.7,
-      "grad_norm": 1.8293100595474243,
+      "grad_norm": 9.149005889892578,
       "learning_rate": 3e-05,
-      "loss": 0.3235,
+      "loss": 0.339,
       "step": 980
     },
     {
       "epoch": 0.7071428571428572,
-      "grad_norm": 0.691096842288971,
+      "grad_norm": 3.723829746246338,
       "learning_rate": 2.9285714285714288e-05,
-      "loss": 0.3713,
+      "loss": 0.3932,
       "step": 990
     },
     {
       "epoch": 0.7142857142857143,
-      "grad_norm": 0.47313550114631653,
+      "grad_norm": 2.2082722187042236,
       "learning_rate": 2.857142857142857e-05,
-      "loss": 0.3652,
+      "loss": 0.3854,
       "step": 1000
     },
     {
       "epoch": 0.7214285714285714,
-      "grad_norm": 0.5191896557807922,
+      "grad_norm": 2.192686080932617,
       "learning_rate": 2.785714285714286e-05,
-      "loss": 0.331,
+      "loss": 0.352,
       "step": 1010
     },
     {
       "epoch": 0.7285714285714285,
-      "grad_norm": 0.509594738483429,
+      "grad_norm": 2.3141751289367676,
       "learning_rate": 2.714285714285714e-05,
-      "loss": 0.2926,
+      "loss": 0.3132,
       "step": 1020
     },
     {
       "epoch": 0.7357142857142858,
-      "grad_norm": 0.9162160754203796,
+      "grad_norm": 4.0575995445251465,
       "learning_rate": 2.642857142857143e-05,
-      "loss": 0.2854,
+      "loss": 0.2929,
       "step": 1030
     },
     {
       "epoch": 0.7428571428571429,
-      "grad_norm": 0.5884422659873962,
+      "grad_norm": 2.5704588890075684,
       "learning_rate": 2.5714285714285714e-05,
-      "loss": 0.3348,
+      "loss": 0.3549,
       "step": 1040
     },
     {
       "epoch": 0.75,
-      "grad_norm": 0.6538064479827881,
+      "grad_norm": 2.970313310623169,
       "learning_rate": 2.5e-05,
-      "loss": 0.2992,
+      "loss": 0.3139,
       "step": 1050
     },
     {
       "epoch": 0.7571428571428571,
-      "grad_norm": 0.49203699827194214,
+      "grad_norm": 3.143388271331787,
       "learning_rate": 2.4285714285714288e-05,
-      "loss": 0.3497,
+      "loss": 0.3732,
       "step": 1060
     },
     {
       "epoch": 0.7642857142857142,
-      "grad_norm": 0.798219621181488,
+      "grad_norm": 3.7672691345214844,
       "learning_rate": 2.357142857142857e-05,
-      "loss": 0.3492,
+      "loss": 0.3752,
       "step": 1070
     },
     {
       "epoch": 0.7714285714285715,
-      "grad_norm": 0.6431849598884583,
+      "grad_norm": 3.106049060821533,
       "learning_rate": 2.2857142857142858e-05,
-      "loss": 0.2953,
+      "loss": 0.3174,
       "step": 1080
     },
     {
       "epoch": 0.7785714285714286,
-      "grad_norm": 0.5676552653312683,
+      "grad_norm": 3.5952601432800293,
       "learning_rate": 2.214285714285714e-05,
-      "loss": 0.323,
+      "loss": 0.3445,
       "step": 1090
     },
     {
       "epoch": 0.7857142857142857,
-      "grad_norm": 0.4670363962650299,
+      "grad_norm": 2.764934778213501,
       "learning_rate": 2.1428571428571428e-05,
-      "loss": 0.3133,
+      "loss": 0.3343,
       "step": 1100
     },
     {
       "epoch": 0.7928571428571428,
-      "grad_norm": 0.6372350454330444,
+      "grad_norm": 3.4910130500793457,
       "learning_rate": 2.0714285714285718e-05,
-      "loss": 0.3025,
+      "loss": 0.3265,
       "step": 1110
     },
     {
       "epoch": 0.8,
-      "grad_norm": 1.2349172830581665,
+      "grad_norm": 6.074836254119873,
       "learning_rate": 2e-05,
-      "loss": 0.3084,
+      "loss": 0.334,
       "step": 1120
     },
     {
       "epoch": 0.8071428571428572,
-      "grad_norm": 0.6304973363876343,
+      "grad_norm": 3.0394229888916016,
       "learning_rate": 1.928571428571429e-05,
-      "loss": 0.2793,
+      "loss": 0.2932,
       "step": 1130
     },
     {
       "epoch": 0.8142857142857143,
-      "grad_norm": 0.47565215826034546,
+      "grad_norm": 2.841371536254883,
       "learning_rate": 1.8571428571428572e-05,
-      "loss": 0.2974,
+      "loss": 0.3143,
       "step": 1140
     },
     {
       "epoch": 0.8214285714285714,
-      "grad_norm": 0.920874297618866,
+      "grad_norm": 3.9328362941741943,
       "learning_rate": 1.785714285714286e-05,
-      "loss": 0.3422,
+      "loss": 0.3726,
       "step": 1150
     },
     {
       "epoch": 0.8285714285714286,
-      "grad_norm": 0.36493098735809326,
+      "grad_norm": 2.6354167461395264,
       "learning_rate": 1.7142857142857145e-05,
-      "loss": 0.3024,
+      "loss": 0.325,
       "step": 1160
     },
     {
       "epoch": 0.8357142857142857,
-      "grad_norm": 0.8869501352310181,
+      "grad_norm": 4.113726615905762,
       "learning_rate": 1.642857142857143e-05,
-      "loss": 0.2979,
+      "loss": 0.3199,
       "step": 1170
     },
     {
       "epoch": 0.8428571428571429,
-      "grad_norm": 0.7538784742355347,
+      "grad_norm": 3.8376033306121826,
       "learning_rate": 1.5714285714285715e-05,
-      "loss": 0.3076,
+      "loss": 0.3299,
       "step": 1180
     },
     {
       "epoch": 0.85,
-      "grad_norm": 0.8670153617858887,
+      "grad_norm": 5.71781063079834,
       "learning_rate": 1.5e-05,
-      "loss": 0.2915,
+      "loss": 0.317,
       "step": 1190
     },
     {
       "epoch": 0.8571428571428571,
-      "grad_norm": 0.6011945009231567,
+      "grad_norm": 3.487872362136841,
       "learning_rate": 1.4285714285714285e-05,
-      "loss": 0.3086,
+      "loss": 0.3301,
       "step": 1200
     },
     {
       "epoch": 0.8642857142857143,
-      "grad_norm": 0.364296555519104,
+      "grad_norm": 1.754288673400879,
       "learning_rate": 1.357142857142857e-05,
-      "loss": 0.3253,
+      "loss": 0.3469,
       "step": 1210
     },
     {
       "epoch": 0.8714285714285714,
-      "grad_norm": 0.7194721102714539,
+      "grad_norm": 3.3238632678985596,
       "learning_rate": 1.2857142857142857e-05,
-      "loss": 0.3211,
+      "loss": 0.3434,
       "step": 1220
     },
     {
       "epoch": 0.8785714285714286,
-      "grad_norm": 0.41066431999206543,
+      "grad_norm": 2.276270627975464,
       "learning_rate": 1.2142857142857144e-05,
-      "loss": 0.2648,
+      "loss": 0.2825,
       "step": 1230
     },
     {
       "epoch": 0.8857142857142857,
-      "grad_norm": 0.7685422301292419,
+      "grad_norm": 4.0892720222473145,
       "learning_rate": 1.1428571428571429e-05,
-      "loss": 0.3156,
+      "loss": 0.3364,
       "step": 1240
     },
     {
       "epoch": 0.8928571428571429,
-      "grad_norm": 0.5432989001274109,
+      "grad_norm": 2.1741929054260254,
       "learning_rate": 1.0714285714285714e-05,
-      "loss": 0.3418,
+      "loss": 0.3592,
       "step": 1250
     },
     {
       "epoch": 0.9,
-      "grad_norm": 0.5999932885169983,
+      "grad_norm": 2.3750619888305664,
       "learning_rate": 1e-05,
-      "loss": 0.381,
+      "loss": 0.4065,
       "step": 1260
     },
     {
       "epoch": 0.9071428571428571,
-      "grad_norm": 0.3930542767047882,
+      "grad_norm": 2.2269678115844727,
       "learning_rate": 9.285714285714286e-06,
-      "loss": 0.384,
+      "loss": 0.4073,
       "step": 1270
     },
     {
       "epoch": 0.9142857142857143,
-      "grad_norm": 0.3380388021469116,
+      "grad_norm": 2.025587320327759,
       "learning_rate": 8.571428571428573e-06,
-      "loss": 0.3303,
+      "loss": 0.3478,
       "step": 1280
     },
     {
       "epoch": 0.9214285714285714,
-      "grad_norm": 1.0583548545837402,
+      "grad_norm": 5.870853900909424,
       "learning_rate": 7.857142857142858e-06,
-      "loss": 0.3573,
+      "loss": 0.386,
       "step": 1290
     },
     {
       "epoch": 0.9285714285714286,
-      "grad_norm": 0.5554153323173523,
+      "grad_norm": 2.9051337242126465,
       "learning_rate": 7.142857142857143e-06,
-      "loss": 0.3045,
+      "loss": 0.3257,
       "step": 1300
     },
     {
       "epoch": 0.9357142857142857,
-      "grad_norm": 0.6635413765907288,
+      "grad_norm": 3.7982888221740723,
       "learning_rate": 6.428571428571429e-06,
-      "loss": 0.3179,
+      "loss": 0.3405,
       "step": 1310
     },
     {
       "epoch": 0.9428571428571428,
-      "grad_norm": 0.9328262805938721,
+      "grad_norm": 5.413133144378662,
       "learning_rate": 5.7142857142857145e-06,
-      "loss": 0.2692,
+      "loss": 0.2909,
       "step": 1320
     },
     {
       "epoch": 0.95,
-      "grad_norm": 0.6629207730293274,
+      "grad_norm": 2.1331217288970947,
       "learning_rate": 5e-06,
-      "loss": 0.3041,
+      "loss": 0.3175,
       "step": 1330
     },
     {
       "epoch": 0.9571428571428572,
-      "grad_norm": 1.1389262676239014,
+      "grad_norm": 4.568922519683838,
       "learning_rate": 4.285714285714286e-06,
-      "loss": 0.321,
+      "loss": 0.3447,
       "step": 1340
     },
     {
       "epoch": 0.9642857142857143,
-      "grad_norm": 0.5038421750068665,
+      "grad_norm": 2.920402765274048,
       "learning_rate": 3.5714285714285714e-06,
-      "loss": 0.3192,
+      "loss": 0.3357,
       "step": 1350
     },
     {
       "epoch": 0.9714285714285714,
-      "grad_norm": 0.4525507986545563,
+      "grad_norm": 1.938262939453125,
       "learning_rate": 2.8571428571428573e-06,
-      "loss": 0.2807,
+      "loss": 0.2942,
       "step": 1360
     },
     {
       "epoch": 0.9785714285714285,
-      "grad_norm": 0.5145785212516785,
+      "grad_norm": 1.8624756336212158,
       "learning_rate": 2.142857142857143e-06,
-      "loss": 0.3307,
+      "loss": 0.3498,
       "step": 1370
     },
     {
       "epoch": 0.9857142857142858,
-      "grad_norm": 0.6031885147094727,
+      "grad_norm": 2.570404291152954,
       "learning_rate": 1.4285714285714286e-06,
-      "loss": 0.3564,
+      "loss": 0.3777,
       "step": 1380
     },
     {
       "epoch": 0.9928571428571429,
-      "grad_norm": 0.38553744554519653,
+      "grad_norm": 1.80099356174469,
       "learning_rate": 7.142857142857143e-07,
-      "loss": 0.315,
+      "loss": 0.3356,
       "step": 1390
     },
     {
       "epoch": 1.0,
-      "grad_norm": 0.5251179933547974,
+      "grad_norm": 2.955940008163452,
       "learning_rate": 0.0,
-      "loss": 0.2818,
+      "loss": 0.3044,
       "step": 1400
     },
     {
       "epoch": 1.0,
-      "eval_loss": 0.29903435707092285,
-      "eval_runtime": 66.066,
-      "eval_samples_per_second": 3.027,
-      "eval_steps_per_second": 0.378,
+      "eval_loss": 0.3177724778652191,
+      "eval_runtime": 67.4116,
+      "eval_samples_per_second": 2.967,
+      "eval_steps_per_second": 0.371,
       "step": 1400
     }
   ],
@@ -1014,7 +1014,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 8487866204160000.0,
+  "total_flos": 8377941884928000.0,
   "train_batch_size": 2,
   "trial_name": null,
   "trial_params": null