Upload folder using huggingface_hub

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:bda320c48ffca2fd27f2dd27592f1c0953ad8c3088f85d6c25100409e26453a5
+oid sha256:d2cd35d0baac967d5cd52ce3fffd9ce200841b230f5ec3f3a91c2d8bca2247d5
 size 3416264

checkpoint-1050/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:bda320c48ffca2fd27f2dd27592f1c0953ad8c3088f85d6c25100409e26453a5
+oid sha256:d2cd35d0baac967d5cd52ce3fffd9ce200841b230f5ec3f3a91c2d8bca2247d5
 size 3416264

checkpoint-1050/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e83eed432c17723901d9349c6d7b3e582ee7aa486717856359ca85d2d4c61f4d
+oid sha256:ae643995a9c608e5fdbb35f7f590c5e913a99f951c4b7e93ba61ed6b84a0f475
 size 6869818

checkpoint-1050/trainer_state.json

@@ -10,761 +10,761 @@
   "log_history": [
     {
       "epoch": 0.02857142857142857,
-      "grad_norm": 11.918978691101074,
+      "grad_norm": 320.9814758300781,
       "learning_rate": 0.0004952380952380952,
-      "loss": 4.4927,
+      "loss": 5.144,
       "step": 10
     },
     {
       "epoch": 0.05714285714285714,
-      "grad_norm": 5.589139461517334,
+      "grad_norm": 8.983612060546875,
       "learning_rate": 0.0004904761904761905,
-      "loss": 2.6308,
+      "loss": 3.5072,
       "step": 20
     },
     {
       "epoch": 0.08571428571428572,
-      "grad_norm": 9.946991920471191,
+      "grad_norm": 5.370372772216797,
       "learning_rate": 0.0004857142857142857,
-      "loss": 2.2656,
+      "loss": 2.2891,
       "step": 30
     },
     {
       "epoch": 0.11428571428571428,
-      "grad_norm": 4.425405502319336,
+      "grad_norm": 2.666455030441284,
       "learning_rate": 0.00048095238095238095,
-      "loss": 2.0816,
+      "loss": 2.0865,
       "step": 40
     },
     {
       "epoch": 0.14285714285714285,
-      "grad_norm": 4.761979103088379,
+      "grad_norm": 4.363830089569092,
       "learning_rate": 0.0004761904761904762,
-      "loss": 2.0915,
+      "loss": 2.061,
       "step": 50
     },
     {
       "epoch": 0.17142857142857143,
-      "grad_norm": 4.236316204071045,
+      "grad_norm": 5.7685394287109375,
       "learning_rate": 0.0004714285714285714,
-      "loss": 1.8267,
+      "loss": 1.8436,
       "step": 60
     },
     {
       "epoch": 0.2,
-      "grad_norm": 3.1871836185455322,
+      "grad_norm": 4.38090181350708,
       "learning_rate": 0.00046666666666666666,
-      "loss": 1.8426,
+      "loss": 1.8594,
       "step": 70
     },
     {
       "epoch": 0.22857142857142856,
-      "grad_norm": 5.339052677154541,
+      "grad_norm": 6.654545783996582,
       "learning_rate": 0.00046190476190476195,
-      "loss": 1.8622,
+      "loss": 1.8804,
       "step": 80
     },
     {
       "epoch": 0.2571428571428571,
-      "grad_norm": 3.1449456214904785,
+      "grad_norm": 2.162074565887451,
       "learning_rate": 0.00045714285714285713,
-      "loss": 1.6587,
+      "loss": 1.6293,
       "step": 90
     },
     {
       "epoch": 0.2857142857142857,
-      "grad_norm": 5.643612861633301,
+      "grad_norm": 4.620967388153076,
       "learning_rate": 0.00045238095238095237,
-      "loss": 1.67,
+      "loss": 1.6454,
       "step": 100
     },
     {
       "epoch": 0.3142857142857143,
-      "grad_norm": 3.7077548503875732,
+      "grad_norm": 5.860279083251953,
       "learning_rate": 0.00044761904761904766,
-      "loss": 1.6768,
+      "loss": 1.6508,
       "step": 110
     },
     {
       "epoch": 0.34285714285714286,
-      "grad_norm": 2.433858633041382,
+      "grad_norm": 2.8689663410186768,
       "learning_rate": 0.00044285714285714284,
-      "loss": 1.3662,
+      "loss": 1.3364,
       "step": 120
     },
     {
       "epoch": 0.37142857142857144,
-      "grad_norm": 3.0955514907836914,
+      "grad_norm": 2.600432872772217,
       "learning_rate": 0.0004380952380952381,
-      "loss": 1.5289,
+      "loss": 1.4913,
       "step": 130
     },
     {
       "epoch": 0.4,
-      "grad_norm": 6.524577617645264,
+      "grad_norm": 3.5309250354766846,
       "learning_rate": 0.00043333333333333337,
-      "loss": 1.4246,
+      "loss": 1.3719,
       "step": 140
     },
     {
       "epoch": 0.42857142857142855,
-      "grad_norm": 4.222738265991211,
+      "grad_norm": 2.634753704071045,
       "learning_rate": 0.00042857142857142855,
-      "loss": 1.5386,
+      "loss": 1.5274,
       "step": 150
     },
     {
       "epoch": 0.45714285714285713,
-      "grad_norm": 3.318093776702881,
+      "grad_norm": 3.284735918045044,
       "learning_rate": 0.0004238095238095238,
-      "loss": 1.6411,
+      "loss": 1.5985,
       "step": 160
     },
     {
       "epoch": 0.4857142857142857,
-      "grad_norm": 3.080537796020508,
+      "grad_norm": 3.2241106033325195,
       "learning_rate": 0.0004190476190476191,
-      "loss": 1.2947,
+      "loss": 1.2673,
       "step": 170
     },
     {
       "epoch": 0.5142857142857142,
-      "grad_norm": 4.210653305053711,
+      "grad_norm": 2.514350175857544,
       "learning_rate": 0.0004142857142857143,
-      "loss": 1.4724,
+      "loss": 1.4211,
       "step": 180
     },
     {
       "epoch": 0.5428571428571428,
-      "grad_norm": 2.615692138671875,
+      "grad_norm": 3.476823091506958,
       "learning_rate": 0.00040952380952380955,
-      "loss": 1.4507,
+      "loss": 1.4254,
       "step": 190
     },
     {
       "epoch": 0.5714285714285714,
-      "grad_norm": 3.874413013458252,
+      "grad_norm": 2.4598169326782227,
       "learning_rate": 0.0004047619047619048,
-      "loss": 1.2955,
+      "loss": 1.2857,
       "step": 200
     },
     {
       "epoch": 0.6,
-      "grad_norm": 4.24971866607666,
+      "grad_norm": 3.915236234664917,
       "learning_rate": 0.0004,
-      "loss": 1.2389,
+      "loss": 1.2144,
       "step": 210
     },
     {
       "epoch": 0.6285714285714286,
-      "grad_norm": 4.347882270812988,
+      "grad_norm": 3.5647623538970947,
       "learning_rate": 0.00039523809523809526,
-      "loss": 1.2475,
+      "loss": 1.2258,
       "step": 220
     },
     {
       "epoch": 0.6571428571428571,
-      "grad_norm": 4.45231819152832,
+      "grad_norm": 4.019155025482178,
       "learning_rate": 0.0003904761904761905,
-      "loss": 1.4072,
+      "loss": 1.4103,
       "step": 230
     },
     {
       "epoch": 0.6857142857142857,
-      "grad_norm": 2.5132458209991455,
+      "grad_norm": 3.1268930435180664,
       "learning_rate": 0.0003857142857142857,
-      "loss": 1.2506,
+      "loss": 1.2308,
       "step": 240
     },
     {
       "epoch": 0.7142857142857143,
-      "grad_norm": 2.3154046535491943,
+      "grad_norm": 2.708326578140259,
       "learning_rate": 0.00038095238095238096,
-      "loss": 1.3371,
+      "loss": 1.3197,
       "step": 250
     },
     {
       "epoch": 0.7428571428571429,
-      "grad_norm": 2.9877989292144775,
+      "grad_norm": 2.4128565788269043,
       "learning_rate": 0.0003761904761904762,
-      "loss": 1.2249,
+      "loss": 1.1922,
       "step": 260
     },
     {
       "epoch": 0.7714285714285715,
-      "grad_norm": 4.407787799835205,
+      "grad_norm": 4.487890243530273,
       "learning_rate": 0.00037142857142857143,
-      "loss": 1.2712,
+      "loss": 1.2505,
       "step": 270
     },
     {
       "epoch": 0.8,
-      "grad_norm": 2.691910982131958,
+      "grad_norm": 3.1995606422424316,
       "learning_rate": 0.00036666666666666667,
-      "loss": 1.2366,
+      "loss": 1.2047,
       "step": 280
     },
     {
       "epoch": 0.8285714285714286,
-      "grad_norm": 2.3351523876190186,
+      "grad_norm": 3.0884156227111816,
       "learning_rate": 0.0003619047619047619,
-      "loss": 1.1777,
+      "loss": 1.1397,
       "step": 290
     },
     {
       "epoch": 0.8571428571428571,
-      "grad_norm": 2.878443956375122,
+      "grad_norm": 3.9434640407562256,
       "learning_rate": 0.00035714285714285714,
-      "loss": 1.1375,
+      "loss": 1.0844,
       "step": 300
     },
     {
       "epoch": 0.8857142857142857,
-      "grad_norm": 3.0720036029815674,
+      "grad_norm": 3.6582884788513184,
       "learning_rate": 0.00035238095238095243,
-      "loss": 1.1578,
+      "loss": 1.1461,
       "step": 310
     },
     {
       "epoch": 0.9142857142857143,
-      "grad_norm": 2.359656572341919,
+      "grad_norm": 2.6335928440093994,
       "learning_rate": 0.0003476190476190476,
-      "loss": 1.4001,
+      "loss": 1.4124,
       "step": 320
     },
     {
       "epoch": 0.9428571428571428,
-      "grad_norm": 2.3982980251312256,
+      "grad_norm": 2.8153300285339355,
       "learning_rate": 0.00034285714285714285,
-      "loss": 1.1716,
+      "loss": 1.1619,
       "step": 330
     },
     {
       "epoch": 0.9714285714285714,
-      "grad_norm": 4.372157573699951,
+      "grad_norm": 5.384872913360596,
       "learning_rate": 0.00033809523809523814,
-      "loss": 1.1177,
+      "loss": 1.1091,
       "step": 340
     },
     {
       "epoch": 1.0,
-      "grad_norm": 4.467468738555908,
+      "grad_norm": 4.4129958152771,
       "learning_rate": 0.0003333333333333333,
-      "loss": 1.1884,
+      "loss": 1.1665,
       "step": 350
     },
     {
       "epoch": 1.0,
-      "eval_loss": 1.0876593589782715,
-      "eval_runtime": 16.2651,
-      "eval_samples_per_second": 12.296,
-      "eval_steps_per_second": 1.537,
+      "eval_loss": 1.0720068216323853,
+      "eval_runtime": 16.9829,
+      "eval_samples_per_second": 11.777,
+      "eval_steps_per_second": 1.472,
       "step": 350
     },
     {
       "epoch": 1.0285714285714285,
-      "grad_norm": 8.64857292175293,
+      "grad_norm": 7.848823547363281,
       "learning_rate": 0.00032857142857142856,
-      "loss": 1.109,
+      "loss": 1.073,
       "step": 360
     },
     {
       "epoch": 1.0571428571428572,
-      "grad_norm": 3.4350078105926514,
+      "grad_norm": 3.559180498123169,
       "learning_rate": 0.00032380952380952385,
-      "loss": 1.2132,
+      "loss": 1.1773,
       "step": 370
     },
     {
       "epoch": 1.0857142857142856,
-      "grad_norm": 2.4915060997009277,
+      "grad_norm": 2.477151870727539,
       "learning_rate": 0.00031904761904761903,
-      "loss": 1.0756,
+      "loss": 1.0647,
       "step": 380
     },
     {
       "epoch": 1.1142857142857143,
-      "grad_norm": 3.537388324737549,
+      "grad_norm": 3.4121389389038086,
       "learning_rate": 0.00031428571428571427,
-      "loss": 1.0506,
+      "loss": 1.0428,
       "step": 390
     },
     {
       "epoch": 1.1428571428571428,
-      "grad_norm": 3.6561241149902344,
+      "grad_norm": 6.0573015213012695,
       "learning_rate": 0.00030952380952380956,
-      "loss": 1.1051,
+      "loss": 1.0667,
       "step": 400
     },
     {
       "epoch": 1.1714285714285715,
-      "grad_norm": 3.6229779720306396,
+      "grad_norm": 3.7485954761505127,
       "learning_rate": 0.0003047619047619048,
-      "loss": 1.0663,
+      "loss": 1.0497,
       "step": 410
     },
     {
       "epoch": 1.2,
-      "grad_norm": 4.541111946105957,
+      "grad_norm": 4.239710330963135,
       "learning_rate": 0.0003,
-      "loss": 1.1192,
+      "loss": 1.1227,
       "step": 420
     },
     {
       "epoch": 1.2285714285714286,
-      "grad_norm": 8.634881973266602,
+      "grad_norm": 7.383611679077148,
       "learning_rate": 0.00029523809523809526,
-      "loss": 1.042,
+      "loss": 1.0,
       "step": 430
     },
     {
       "epoch": 1.2571428571428571,
-      "grad_norm": 2.93104887008667,
+      "grad_norm": 3.1444430351257324,
       "learning_rate": 0.0002904761904761905,
-      "loss": 0.9288,
+      "loss": 0.9194,
       "step": 440
     },
     {
       "epoch": 1.2857142857142856,
-      "grad_norm": 6.468315601348877,
+      "grad_norm": 8.754597663879395,
       "learning_rate": 0.0002857142857142857,
-      "loss": 1.2407,
+      "loss": 1.2234,
       "step": 450
     },
     {
       "epoch": 1.3142857142857143,
-      "grad_norm": 5.183238506317139,
+      "grad_norm": 5.383545875549316,
       "learning_rate": 0.00028095238095238097,
-      "loss": 1.0302,
+      "loss": 1.0219,
       "step": 460
     },
     {
       "epoch": 1.342857142857143,
-      "grad_norm": 2.7399845123291016,
+      "grad_norm": 3.977374792098999,
       "learning_rate": 0.0002761904761904762,
-      "loss": 1.096,
+      "loss": 1.061,
       "step": 470
     },
     {
       "epoch": 1.3714285714285714,
-      "grad_norm": 4.08115291595459,
+      "grad_norm": 3.6732325553894043,
       "learning_rate": 0.0002714285714285714,
-      "loss": 1.0931,
+      "loss": 1.1131,
       "step": 480
     },
     {
       "epoch": 1.4,
-      "grad_norm": 3.567685842514038,
+      "grad_norm": 4.493168830871582,
       "learning_rate": 0.0002666666666666667,
-      "loss": 0.9909,
+      "loss": 1.0033,
       "step": 490
     },
     {
       "epoch": 1.4285714285714286,
-      "grad_norm": 3.2980871200561523,
+      "grad_norm": 2.9993937015533447,
       "learning_rate": 0.0002619047619047619,
-      "loss": 1.0007,
+      "loss": 1.0023,
       "step": 500
     },
     {
       "epoch": 1.457142857142857,
-      "grad_norm": 4.056380271911621,
+      "grad_norm": 3.984879732131958,
       "learning_rate": 0.0002571428571428571,
-      "loss": 0.9869,
+      "loss": 0.9796,
       "step": 510
     },
     {
       "epoch": 1.4857142857142858,
-      "grad_norm": 5.267114162445068,
+      "grad_norm": 5.652857303619385,
       "learning_rate": 0.0002523809523809524,
-      "loss": 0.9149,
+      "loss": 0.912,
       "step": 520
     },
     {
       "epoch": 1.5142857142857142,
-      "grad_norm": 3.0776150226593018,
+      "grad_norm": 4.275585174560547,
       "learning_rate": 0.0002476190476190476,
-      "loss": 0.9842,
+      "loss": 0.9618,
       "step": 530
     },
     {
       "epoch": 1.5428571428571427,
-      "grad_norm": 3.3917739391326904,
+      "grad_norm": 4.000895977020264,
       "learning_rate": 0.00024285714285714286,
-      "loss": 1.0165,
+      "loss": 1.0019,
       "step": 540
     },
     {
       "epoch": 1.5714285714285714,
-      "grad_norm": 3.262791633605957,
+      "grad_norm": 3.681532144546509,
       "learning_rate": 0.0002380952380952381,
-      "loss": 0.9766,
+      "loss": 0.9769,
       "step": 550
     },
     {
       "epoch": 1.6,
-      "grad_norm": 5.334137916564941,
+      "grad_norm": 6.31843376159668,
       "learning_rate": 0.00023333333333333333,
-      "loss": 1.0761,
+      "loss": 1.0508,
       "step": 560
     },
     {
       "epoch": 1.6285714285714286,
-      "grad_norm": 3.0354158878326416,
+      "grad_norm": 3.165881633758545,
       "learning_rate": 0.00022857142857142857,
-      "loss": 1.1283,
+      "loss": 1.1192,
       "step": 570
     },
     {
       "epoch": 1.657142857142857,
-      "grad_norm": 2.9675745964050293,
+      "grad_norm": 3.3488543033599854,
       "learning_rate": 0.00022380952380952383,
-      "loss": 0.9564,
+      "loss": 0.945,
       "step": 580
     },
     {
       "epoch": 1.6857142857142857,
-      "grad_norm": 3.70650053024292,
+      "grad_norm": 3.479367256164551,
       "learning_rate": 0.00021904761904761904,
-      "loss": 1.1382,
+      "loss": 1.1191,
       "step": 590
     },
     {
       "epoch": 1.7142857142857144,
-      "grad_norm": 4.117910385131836,
+      "grad_norm": 3.9680163860321045,
       "learning_rate": 0.00021428571428571427,
-      "loss": 1.0104,
+      "loss": 1.0116,
       "step": 600
     },
     {
       "epoch": 1.7428571428571429,
-      "grad_norm": 3.287292957305908,
+      "grad_norm": 3.4398531913757324,
       "learning_rate": 0.00020952380952380954,
-      "loss": 1.0152,
+      "loss": 1.0162,
       "step": 610
     },
     {
       "epoch": 1.7714285714285714,
-      "grad_norm": 4.567937850952148,
+      "grad_norm": 4.210978031158447,
       "learning_rate": 0.00020476190476190477,
-      "loss": 0.9348,
+      "loss": 0.9214,
       "step": 620
     },
     {
       "epoch": 1.8,
-      "grad_norm": 3.143092393875122,
+      "grad_norm": 3.9029576778411865,
       "learning_rate": 0.0002,
-      "loss": 1.0373,
+      "loss": 1.0135,
       "step": 630
     },
     {
       "epoch": 1.8285714285714287,
-      "grad_norm": 3.003242015838623,
+      "grad_norm": 3.1066203117370605,
       "learning_rate": 0.00019523809523809525,
-      "loss": 1.0157,
+      "loss": 1.0136,
       "step": 640
     },
     {
       "epoch": 1.8571428571428572,
-      "grad_norm": 5.4209489822387695,
+      "grad_norm": 4.775870323181152,
       "learning_rate": 0.00019047619047619048,
-      "loss": 0.8729,
+      "loss": 0.8558,
       "step": 650
     },
     {
       "epoch": 1.8857142857142857,
-      "grad_norm": 8.428689956665039,
+      "grad_norm": 4.7907490730285645,
       "learning_rate": 0.00018571428571428572,
-      "loss": 0.9703,
+      "loss": 0.9436,
       "step": 660
     },
     {
       "epoch": 1.9142857142857141,
-      "grad_norm": 4.8756866455078125,
+      "grad_norm": 4.235854148864746,
       "learning_rate": 0.00018095238095238095,
-      "loss": 1.0013,
+      "loss": 0.9884,
       "step": 670
     },
     {
       "epoch": 1.9428571428571428,
-      "grad_norm": 3.252725601196289,
+      "grad_norm": 3.239032030105591,
       "learning_rate": 0.00017619047619047622,
-      "loss": 0.9414,
+      "loss": 0.9304,
       "step": 680
     },
     {
       "epoch": 1.9714285714285715,
-      "grad_norm": 3.5052497386932373,
+      "grad_norm": 3.5313665866851807,
       "learning_rate": 0.00017142857142857143,
-      "loss": 0.9891,
+      "loss": 0.983,
       "step": 690
     },
     {
       "epoch": 2.0,
-      "grad_norm": 2.886568546295166,
+      "grad_norm": 3.0609097480773926,
       "learning_rate": 0.00016666666666666666,
-      "loss": 1.0002,
+      "loss": 0.9815,
       "step": 700
     },
     {
       "epoch": 2.0,
-      "eval_loss": 0.9278395175933838,
-      "eval_runtime": 16.2518,
-      "eval_samples_per_second": 12.306,
-      "eval_steps_per_second": 1.538,
+      "eval_loss": 0.8912971615791321,
+      "eval_runtime": 17.0248,
+      "eval_samples_per_second": 11.748,
+      "eval_steps_per_second": 1.468,
       "step": 700
     },
     {
       "epoch": 2.0285714285714285,
-      "grad_norm": 2.525127649307251,
+      "grad_norm": 2.448066473007202,
       "learning_rate": 0.00016190476190476192,
-      "loss": 0.8582,
+      "loss": 0.8429,
       "step": 710
     },
     {
       "epoch": 2.057142857142857,
-      "grad_norm": 2.341196298599243,
+      "grad_norm": 2.44789719581604,
       "learning_rate": 0.00015714285714285713,
-      "loss": 0.8304,
+      "loss": 0.8165,
       "step": 720
     },
     {
       "epoch": 2.085714285714286,
-      "grad_norm": 2.54571795463562,
+      "grad_norm": 3.3588461875915527,
       "learning_rate": 0.0001523809523809524,
-      "loss": 0.8242,
+      "loss": 0.8081,
       "step": 730
     },
     {
       "epoch": 2.1142857142857143,
-      "grad_norm": 3.9072329998016357,
+      "grad_norm": 3.647935628890991,
       "learning_rate": 0.00014761904761904763,
-      "loss": 0.8259,
+      "loss": 0.8373,
       "step": 740
     },
     {
       "epoch": 2.142857142857143,
-      "grad_norm": 5.519943714141846,
+      "grad_norm": 5.533905506134033,
       "learning_rate": 0.00014285714285714284,
-      "loss": 1.0519,
+      "loss": 1.0447,
       "step": 750
     },
     {
       "epoch": 2.1714285714285713,
-      "grad_norm": 3.61736798286438,
+      "grad_norm": 3.834791660308838,
       "learning_rate": 0.0001380952380952381,
-      "loss": 0.8888,
+      "loss": 0.909,
       "step": 760
     },
     {
       "epoch": 2.2,
-      "grad_norm": 3.1113357543945312,
+      "grad_norm": 3.3422257900238037,
       "learning_rate": 0.00013333333333333334,
-      "loss": 0.8582,
+      "loss": 0.8473,
       "step": 770
     },
     {
       "epoch": 2.2285714285714286,
-      "grad_norm": 4.208303451538086,
+      "grad_norm": 3.9888529777526855,
       "learning_rate": 0.00012857142857142855,
-      "loss": 0.9397,
+      "loss": 0.9289,
       "step": 780
     },
     {
       "epoch": 2.257142857142857,
-      "grad_norm": 4.2059454917907715,
+      "grad_norm": 2.7921173572540283,
       "learning_rate": 0.0001238095238095238,
-      "loss": 0.7807,
+      "loss": 0.7639,
       "step": 790
     },
     {
       "epoch": 2.2857142857142856,
-      "grad_norm": 3.725945234298706,
+      "grad_norm": 3.649446964263916,
       "learning_rate": 0.00011904761904761905,
-      "loss": 0.8144,
+      "loss": 0.8032,
       "step": 800
     },
     {
       "epoch": 2.314285714285714,
-      "grad_norm": 3.4127461910247803,
+      "grad_norm": 2.795527219772339,
       "learning_rate": 0.00011428571428571428,
-      "loss": 0.8632,
+      "loss": 0.8604,
       "step": 810
     },
     {
       "epoch": 2.342857142857143,
-      "grad_norm": 3.4482555389404297,
+      "grad_norm": 4.063432693481445,
       "learning_rate": 0.00010952380952380952,
-      "loss": 0.8148,
+      "loss": 0.8038,
       "step": 820
     },
     {
       "epoch": 2.3714285714285714,
-      "grad_norm": 3.38759183883667,
+      "grad_norm": 6.527313232421875,
       "learning_rate": 0.00010476190476190477,
-      "loss": 0.9791,
+      "loss": 0.9626,
       "step": 830
     },
     {
       "epoch": 2.4,
-      "grad_norm": 4.449093818664551,
+      "grad_norm": 4.3589677810668945,
       "learning_rate": 0.0001,
-      "loss": 0.8427,
+      "loss": 0.8209,
       "step": 840
     },
     {
       "epoch": 2.4285714285714284,
-      "grad_norm": 6.127928256988525,
+      "grad_norm": 4.879549026489258,
       "learning_rate": 9.523809523809524e-05,
-      "loss": 0.7613,
+      "loss": 0.7638,
       "step": 850
     },
     {
       "epoch": 2.4571428571428573,
-      "grad_norm": 4.715824604034424,
+      "grad_norm": 4.141664981842041,
       "learning_rate": 9.047619047619048e-05,
-      "loss": 0.9275,
+      "loss": 0.8789,
       "step": 860
     },
     {
       "epoch": 2.4857142857142858,
-      "grad_norm": 2.2265524864196777,
+      "grad_norm": 2.2482237815856934,
       "learning_rate": 8.571428571428571e-05,
-      "loss": 0.9554,
+      "loss": 0.9195,
       "step": 870
     },
     {
       "epoch": 2.5142857142857142,
-      "grad_norm": 3.3588171005249023,
+      "grad_norm": 3.0567550659179688,
       "learning_rate": 8.095238095238096e-05,
-      "loss": 0.8616,
+      "loss": 0.8484,
       "step": 880
     },
     {
       "epoch": 2.5428571428571427,
-      "grad_norm": 4.424010753631592,
+      "grad_norm": 4.093011379241943,
       "learning_rate": 7.61904761904762e-05,
-      "loss": 0.9946,
+      "loss": 0.9751,
       "step": 890
     },
     {
       "epoch": 2.571428571428571,
-      "grad_norm": 5.702920913696289,
+      "grad_norm": 4.998325824737549,
       "learning_rate": 7.142857142857142e-05,
-      "loss": 0.6807,
+      "loss": 0.6668,
       "step": 900
     },
     {
       "epoch": 2.6,
-      "grad_norm": 2.8263401985168457,
+      "grad_norm": 2.410688877105713,
       "learning_rate": 6.666666666666667e-05,
-      "loss": 0.8335,
+      "loss": 0.8172,
       "step": 910
     },
     {
       "epoch": 2.6285714285714286,
-      "grad_norm": 4.067463397979736,
+      "grad_norm": 4.06935453414917,
       "learning_rate": 6.19047619047619e-05,
-      "loss": 0.9006,
+      "loss": 0.8715,
       "step": 920
     },
     {
       "epoch": 2.657142857142857,
-      "grad_norm": 4.599573612213135,
+      "grad_norm": 4.4172749519348145,
       "learning_rate": 5.714285714285714e-05,
-      "loss": 0.7835,
+      "loss": 0.7697,
       "step": 930
     },
     {
       "epoch": 2.685714285714286,
-      "grad_norm": 4.713394641876221,
+      "grad_norm": 4.751075267791748,
       "learning_rate": 5.2380952380952384e-05,
-      "loss": 0.9429,
+      "loss": 0.9336,
       "step": 940
     },
     {
       "epoch": 2.7142857142857144,
-      "grad_norm": 4.743385314941406,
+      "grad_norm": 5.419342994689941,
       "learning_rate": 4.761904761904762e-05,
-      "loss": 0.7886,
+      "loss": 0.7864,
       "step": 950
     },
     {
       "epoch": 2.742857142857143,
-      "grad_norm": 4.933552265167236,
+      "grad_norm": 5.93941593170166,
       "learning_rate": 4.2857142857142856e-05,
-      "loss": 0.7563,
+      "loss": 0.7512,
       "step": 960
     },
     {
       "epoch": 2.7714285714285714,
-      "grad_norm": 4.732188701629639,
+      "grad_norm": 4.589322090148926,
       "learning_rate": 3.80952380952381e-05,
-      "loss": 0.9169,
+      "loss": 0.931,
       "step": 970
     },
     {
       "epoch": 2.8,
-      "grad_norm": 2.9946858882904053,
+      "grad_norm": 2.8117074966430664,
       "learning_rate": 3.3333333333333335e-05,
-      "loss": 0.9292,
+      "loss": 0.9269,
       "step": 980
     },
     {
       "epoch": 2.8285714285714287,
-      "grad_norm": 3.2630443572998047,
+      "grad_norm": 3.5889601707458496,
       "learning_rate": 2.857142857142857e-05,
-      "loss": 0.8928,
+      "loss": 0.8692,
       "step": 990
     },
     {
       "epoch": 2.857142857142857,
-      "grad_norm": 3.2443957328796387,
+      "grad_norm": 2.896406412124634,
       "learning_rate": 2.380952380952381e-05,
-      "loss": 0.8818,
+      "loss": 0.8653,
       "step": 1000
     },
     {
       "epoch": 2.8857142857142857,
-      "grad_norm": 4.648172378540039,
+      "grad_norm": 4.262946605682373,
       "learning_rate": 1.904761904761905e-05,
-      "loss": 0.8155,
+      "loss": 0.8147,
       "step": 1010
     },
     {
       "epoch": 2.914285714285714,
-      "grad_norm": 2.9092743396759033,
+      "grad_norm": 2.0895347595214844,
       "learning_rate": 1.4285714285714285e-05,
-      "loss": 0.7833,
+      "loss": 0.7894,
       "step": 1020
     },
     {
       "epoch": 2.942857142857143,
-      "grad_norm": 4.052741050720215,
+      "grad_norm": 3.1296284198760986,
       "learning_rate": 9.523809523809525e-06,
-      "loss": 0.6906,
+      "loss": 0.6825,
       "step": 1030
     },
     {
       "epoch": 2.9714285714285715,
-      "grad_norm": 4.878310680389404,
+      "grad_norm": 4.006032943725586,
       "learning_rate": 4.761904761904762e-06,
-      "loss": 0.6979,
+      "loss": 0.7002,
       "step": 1040
     },
     {
       "epoch": 3.0,
-      "grad_norm": 4.789093971252441,
+      "grad_norm": 4.667923450469971,
       "learning_rate": 0.0,
-      "loss": 0.8579,
+      "loss": 0.854,
       "step": 1050
     },
     {
       "epoch": 3.0,
-      "eval_loss": 0.8077505230903625,
-      "eval_runtime": 16.2781,
-      "eval_samples_per_second": 12.286,
-      "eval_steps_per_second": 1.536,
+      "eval_loss": 0.7869462370872498,
+      "eval_runtime": 16.9799,
+      "eval_samples_per_second": 11.779,
+      "eval_steps_per_second": 1.472,
       "step": 1050
     }
   ],

checkpoint-350/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3d7c63baa562d4afaa4d411aa139867d68a941e5ff4c311e3978a2a1b4b34fe2
+oid sha256:07b19a3342e5274099483d3b96acb4e8a3b702bfff8b05bd21955b7225a6d5f3
 size 3416264

checkpoint-350/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6862b2082074ac04ffc613d6b1c3a96957b5d858c58de54d79f56adb2d732eb6
+oid sha256:30c9044fc476d6d5c60a711eb88687b34e285486823fc0794f9256ea8ee19d8f
 size 6869818

checkpoint-350/trainer_state.json

@@ -10,255 +10,255 @@
   "log_history": [
     {
       "epoch": 0.02857142857142857,
-      "grad_norm": 11.918978691101074,
+      "grad_norm": 320.9814758300781,
       "learning_rate": 0.0004952380952380952,
-      "loss": 4.4927,
+      "loss": 5.144,
       "step": 10
     },
     {
       "epoch": 0.05714285714285714,
-      "grad_norm": 5.589139461517334,
+      "grad_norm": 8.983612060546875,
       "learning_rate": 0.0004904761904761905,
-      "loss": 2.6308,
+      "loss": 3.5072,
       "step": 20
     },
     {
       "epoch": 0.08571428571428572,
-      "grad_norm": 9.946991920471191,
+      "grad_norm": 5.370372772216797,
       "learning_rate": 0.0004857142857142857,
-      "loss": 2.2656,
+      "loss": 2.2891,
       "step": 30
     },
     {
       "epoch": 0.11428571428571428,
-      "grad_norm": 4.425405502319336,
+      "grad_norm": 2.666455030441284,
       "learning_rate": 0.00048095238095238095,
-      "loss": 2.0816,
+      "loss": 2.0865,
       "step": 40
     },
     {
       "epoch": 0.14285714285714285,
-      "grad_norm": 4.761979103088379,
+      "grad_norm": 4.363830089569092,
       "learning_rate": 0.0004761904761904762,
-      "loss": 2.0915,
+      "loss": 2.061,
       "step": 50
     },
     {
       "epoch": 0.17142857142857143,
-      "grad_norm": 4.236316204071045,
+      "grad_norm": 5.7685394287109375,
       "learning_rate": 0.0004714285714285714,
-      "loss": 1.8267,
+      "loss": 1.8436,
       "step": 60
     },
     {
       "epoch": 0.2,
-      "grad_norm": 3.1871836185455322,
+      "grad_norm": 4.38090181350708,
       "learning_rate": 0.00046666666666666666,
-      "loss": 1.8426,
+      "loss": 1.8594,
       "step": 70
     },
     {
       "epoch": 0.22857142857142856,
-      "grad_norm": 5.339052677154541,
+      "grad_norm": 6.654545783996582,
       "learning_rate": 0.00046190476190476195,
-      "loss": 1.8622,
+      "loss": 1.8804,
       "step": 80
     },
     {
       "epoch": 0.2571428571428571,
-      "grad_norm": 3.1449456214904785,
+      "grad_norm": 2.162074565887451,
       "learning_rate": 0.00045714285714285713,
-      "loss": 1.6587,
+      "loss": 1.6293,
       "step": 90
     },
     {
       "epoch": 0.2857142857142857,
-      "grad_norm": 5.643612861633301,
+      "grad_norm": 4.620967388153076,
       "learning_rate": 0.00045238095238095237,
-      "loss": 1.67,
+      "loss": 1.6454,
       "step": 100
     },
     {
       "epoch": 0.3142857142857143,
-      "grad_norm": 3.7077548503875732,
+      "grad_norm": 5.860279083251953,
       "learning_rate": 0.00044761904761904766,
-      "loss": 1.6768,
+      "loss": 1.6508,
       "step": 110
     },
     {
       "epoch": 0.34285714285714286,
-      "grad_norm": 2.433858633041382,
+      "grad_norm": 2.8689663410186768,
       "learning_rate": 0.00044285714285714284,
-      "loss": 1.3662,
+      "loss": 1.3364,
       "step": 120
     },
     {
       "epoch": 0.37142857142857144,
-      "grad_norm": 3.0955514907836914,
+      "grad_norm": 2.600432872772217,
       "learning_rate": 0.0004380952380952381,
-      "loss": 1.5289,
+      "loss": 1.4913,
       "step": 130
     },
     {
       "epoch": 0.4,
-      "grad_norm": 6.524577617645264,
+      "grad_norm": 3.5309250354766846,
       "learning_rate": 0.00043333333333333337,
-      "loss": 1.4246,
+      "loss": 1.3719,
       "step": 140
     },
     {
       "epoch": 0.42857142857142855,
-      "grad_norm": 4.222738265991211,
+      "grad_norm": 2.634753704071045,
       "learning_rate": 0.00042857142857142855,
-      "loss": 1.5386,
+      "loss": 1.5274,
       "step": 150
     },
     {
       "epoch": 0.45714285714285713,
-      "grad_norm": 3.318093776702881,
+      "grad_norm": 3.284735918045044,
       "learning_rate": 0.0004238095238095238,
-      "loss": 1.6411,
+      "loss": 1.5985,
       "step": 160
     },
     {
       "epoch": 0.4857142857142857,
-      "grad_norm": 3.080537796020508,
+      "grad_norm": 3.2241106033325195,
       "learning_rate": 0.0004190476190476191,
-      "loss": 1.2947,
+      "loss": 1.2673,
       "step": 170
     },
     {
       "epoch": 0.5142857142857142,
-      "grad_norm": 4.210653305053711,
+      "grad_norm": 2.514350175857544,
       "learning_rate": 0.0004142857142857143,
-      "loss": 1.4724,
+      "loss": 1.4211,
       "step": 180
     },
     {
       "epoch": 0.5428571428571428,
-      "grad_norm": 2.615692138671875,
+      "grad_norm": 3.476823091506958,
       "learning_rate": 0.00040952380952380955,
-      "loss": 1.4507,
+      "loss": 1.4254,
       "step": 190
     },
     {
       "epoch": 0.5714285714285714,
-      "grad_norm": 3.874413013458252,
+      "grad_norm": 2.4598169326782227,
       "learning_rate": 0.0004047619047619048,
-      "loss": 1.2955,
+      "loss": 1.2857,
       "step": 200
     },
     {
       "epoch": 0.6,
-      "grad_norm": 4.24971866607666,
+      "grad_norm": 3.915236234664917,
       "learning_rate": 0.0004,
-      "loss": 1.2389,
+      "loss": 1.2144,
       "step": 210
     },
     {
       "epoch": 0.6285714285714286,
-      "grad_norm": 4.347882270812988,
+      "grad_norm": 3.5647623538970947,
       "learning_rate": 0.00039523809523809526,
-      "loss": 1.2475,
+      "loss": 1.2258,
       "step": 220
     },
     {
       "epoch": 0.6571428571428571,
-      "grad_norm": 4.45231819152832,
+      "grad_norm": 4.019155025482178,
       "learning_rate": 0.0003904761904761905,
-      "loss": 1.4072,
+      "loss": 1.4103,
       "step": 230
     },
     {
       "epoch": 0.6857142857142857,
-      "grad_norm": 2.5132458209991455,
+      "grad_norm": 3.1268930435180664,
       "learning_rate": 0.0003857142857142857,
-      "loss": 1.2506,
+      "loss": 1.2308,
       "step": 240
     },
     {
       "epoch": 0.7142857142857143,
-      "grad_norm": 2.3154046535491943,
+      "grad_norm": 2.708326578140259,
       "learning_rate": 0.00038095238095238096,
-      "loss": 1.3371,
+      "loss": 1.3197,
       "step": 250
     },
     {
       "epoch": 0.7428571428571429,
-      "grad_norm": 2.9877989292144775,
+      "grad_norm": 2.4128565788269043,
       "learning_rate": 0.0003761904761904762,
-      "loss": 1.2249,
+      "loss": 1.1922,
       "step": 260
     },
     {
       "epoch": 0.7714285714285715,
-      "grad_norm": 4.407787799835205,
+      "grad_norm": 4.487890243530273,
       "learning_rate": 0.00037142857142857143,
-      "loss": 1.2712,
+      "loss": 1.2505,
       "step": 270
     },
     {
       "epoch": 0.8,
-      "grad_norm": 2.691910982131958,
+      "grad_norm": 3.1995606422424316,
       "learning_rate": 0.00036666666666666667,
-      "loss": 1.2366,
+      "loss": 1.2047,
       "step": 280
     },
     {
       "epoch": 0.8285714285714286,
-      "grad_norm": 2.3351523876190186,
+      "grad_norm": 3.0884156227111816,
       "learning_rate": 0.0003619047619047619,
-      "loss": 1.1777,
+      "loss": 1.1397,
       "step": 290
     },
     {
       "epoch": 0.8571428571428571,
-      "grad_norm": 2.878443956375122,
+      "grad_norm": 3.9434640407562256,
       "learning_rate": 0.00035714285714285714,
-      "loss": 1.1375,
+      "loss": 1.0844,
       "step": 300
     },
     {
       "epoch": 0.8857142857142857,
-      "grad_norm": 3.0720036029815674,
+      "grad_norm": 3.6582884788513184,
       "learning_rate": 0.00035238095238095243,
-      "loss": 1.1578,
+      "loss": 1.1461,
       "step": 310
     },
     {
       "epoch": 0.9142857142857143,
-      "grad_norm": 2.359656572341919,
+      "grad_norm": 2.6335928440093994,
       "learning_rate": 0.0003476190476190476,
-      "loss": 1.4001,
+      "loss": 1.4124,
       "step": 320
     },
     {
       "epoch": 0.9428571428571428,
-      "grad_norm": 2.3982980251312256,
+      "grad_norm": 2.8153300285339355,
       "learning_rate": 0.00034285714285714285,
-      "loss": 1.1716,
+      "loss": 1.1619,
       "step": 330
     },
     {
       "epoch": 0.9714285714285714,
-      "grad_norm": 4.372157573699951,
+      "grad_norm": 5.384872913360596,
       "learning_rate": 0.00033809523809523814,
-      "loss": 1.1177,
+      "loss": 1.1091,
       "step": 340
     },
     {
       "epoch": 1.0,
-      "grad_norm": 4.467468738555908,
+      "grad_norm": 4.4129958152771,
       "learning_rate": 0.0003333333333333333,
-      "loss": 1.1884,
+      "loss": 1.1665,
       "step": 350
     },
     {
       "epoch": 1.0,
-      "eval_loss": 1.0876593589782715,
-      "eval_runtime": 16.2651,
-      "eval_samples_per_second": 12.296,
-      "eval_steps_per_second": 1.537,
+      "eval_loss": 1.0720068216323853,
+      "eval_runtime": 16.9829,
+      "eval_samples_per_second": 11.777,
+      "eval_steps_per_second": 1.472,
       "step": 350
     }
   ],

checkpoint-700/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ae947385a52bb213b99caff8cc082f21cabcec22c32556b34f3a9e7afd17fe45
+oid sha256:8221c3fd15b62a6a4bef2a0e388070ff537f3157ea9a6ce45f5aa8cc5d7c70ed
 size 3416264

checkpoint-700/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3234b5048504094f841eae51069d43e7ec592a17538699c7c394ca23cbd41e17
+oid sha256:74872d6416fc8a272ada5a388a703d59001f77a2bc5f613c852a4d74b1c54e49
 size 6869818

checkpoint-700/trainer_state.json

@@ -10,508 +10,508 @@
   "log_history": [
     {
       "epoch": 0.02857142857142857,
-      "grad_norm": 11.918978691101074,
+      "grad_norm": 320.9814758300781,
       "learning_rate": 0.0004952380952380952,
-      "loss": 4.4927,
+      "loss": 5.144,
       "step": 10
     },
     {
       "epoch": 0.05714285714285714,
-      "grad_norm": 5.589139461517334,
+      "grad_norm": 8.983612060546875,
       "learning_rate": 0.0004904761904761905,
-      "loss": 2.6308,
+      "loss": 3.5072,
       "step": 20
     },
     {
       "epoch": 0.08571428571428572,
-      "grad_norm": 9.946991920471191,
+      "grad_norm": 5.370372772216797,
       "learning_rate": 0.0004857142857142857,
-      "loss": 2.2656,
+      "loss": 2.2891,
       "step": 30
     },
     {
       "epoch": 0.11428571428571428,
-      "grad_norm": 4.425405502319336,
+      "grad_norm": 2.666455030441284,
       "learning_rate": 0.00048095238095238095,
-      "loss": 2.0816,
+      "loss": 2.0865,
       "step": 40
     },
     {
       "epoch": 0.14285714285714285,
-      "grad_norm": 4.761979103088379,
+      "grad_norm": 4.363830089569092,
       "learning_rate": 0.0004761904761904762,
-      "loss": 2.0915,
+      "loss": 2.061,
       "step": 50
     },
     {
       "epoch": 0.17142857142857143,
-      "grad_norm": 4.236316204071045,
+      "grad_norm": 5.7685394287109375,
       "learning_rate": 0.0004714285714285714,
-      "loss": 1.8267,
+      "loss": 1.8436,
       "step": 60
     },
     {
       "epoch": 0.2,
-      "grad_norm": 3.1871836185455322,
+      "grad_norm": 4.38090181350708,
       "learning_rate": 0.00046666666666666666,
-      "loss": 1.8426,
+      "loss": 1.8594,
       "step": 70
     },
     {
       "epoch": 0.22857142857142856,
-      "grad_norm": 5.339052677154541,
+      "grad_norm": 6.654545783996582,
       "learning_rate": 0.00046190476190476195,
-      "loss": 1.8622,
+      "loss": 1.8804,
       "step": 80
     },
     {
       "epoch": 0.2571428571428571,
-      "grad_norm": 3.1449456214904785,
+      "grad_norm": 2.162074565887451,
       "learning_rate": 0.00045714285714285713,
-      "loss": 1.6587,
+      "loss": 1.6293,
       "step": 90
     },
     {
       "epoch": 0.2857142857142857,
-      "grad_norm": 5.643612861633301,
+      "grad_norm": 4.620967388153076,
       "learning_rate": 0.00045238095238095237,
-      "loss": 1.67,
+      "loss": 1.6454,
       "step": 100
     },
     {
       "epoch": 0.3142857142857143,
-      "grad_norm": 3.7077548503875732,
+      "grad_norm": 5.860279083251953,
       "learning_rate": 0.00044761904761904766,
-      "loss": 1.6768,
+      "loss": 1.6508,
       "step": 110
     },
     {
       "epoch": 0.34285714285714286,
-      "grad_norm": 2.433858633041382,
+      "grad_norm": 2.8689663410186768,
       "learning_rate": 0.00044285714285714284,
-      "loss": 1.3662,
+      "loss": 1.3364,
       "step": 120
     },
     {
       "epoch": 0.37142857142857144,
-      "grad_norm": 3.0955514907836914,
+      "grad_norm": 2.600432872772217,
       "learning_rate": 0.0004380952380952381,
-      "loss": 1.5289,
+      "loss": 1.4913,
       "step": 130
     },
     {
       "epoch": 0.4,
-      "grad_norm": 6.524577617645264,
+      "grad_norm": 3.5309250354766846,
       "learning_rate": 0.00043333333333333337,
-      "loss": 1.4246,
+      "loss": 1.3719,
       "step": 140
     },
     {
       "epoch": 0.42857142857142855,
-      "grad_norm": 4.222738265991211,
+      "grad_norm": 2.634753704071045,
       "learning_rate": 0.00042857142857142855,
-      "loss": 1.5386,
+      "loss": 1.5274,
       "step": 150
     },
     {
       "epoch": 0.45714285714285713,
-      "grad_norm": 3.318093776702881,
+      "grad_norm": 3.284735918045044,
       "learning_rate": 0.0004238095238095238,
-      "loss": 1.6411,
+      "loss": 1.5985,
       "step": 160
     },
     {
       "epoch": 0.4857142857142857,
-      "grad_norm": 3.080537796020508,
+      "grad_norm": 3.2241106033325195,
       "learning_rate": 0.0004190476190476191,
-      "loss": 1.2947,
+      "loss": 1.2673,
       "step": 170
     },
     {
       "epoch": 0.5142857142857142,
-      "grad_norm": 4.210653305053711,
+      "grad_norm": 2.514350175857544,
       "learning_rate": 0.0004142857142857143,
-      "loss": 1.4724,
+      "loss": 1.4211,
       "step": 180
     },
     {
       "epoch": 0.5428571428571428,
-      "grad_norm": 2.615692138671875,
+      "grad_norm": 3.476823091506958,
       "learning_rate": 0.00040952380952380955,
-      "loss": 1.4507,
+      "loss": 1.4254,
       "step": 190
     },
     {
       "epoch": 0.5714285714285714,
-      "grad_norm": 3.874413013458252,
+      "grad_norm": 2.4598169326782227,
       "learning_rate": 0.0004047619047619048,
-      "loss": 1.2955,
+      "loss": 1.2857,
       "step": 200
     },
     {
       "epoch": 0.6,
-      "grad_norm": 4.24971866607666,
+      "grad_norm": 3.915236234664917,
       "learning_rate": 0.0004,
-      "loss": 1.2389,
+      "loss": 1.2144,
       "step": 210
     },
     {
       "epoch": 0.6285714285714286,
-      "grad_norm": 4.347882270812988,
+      "grad_norm": 3.5647623538970947,
       "learning_rate": 0.00039523809523809526,
-      "loss": 1.2475,
+      "loss": 1.2258,
       "step": 220
     },
     {
       "epoch": 0.6571428571428571,
-      "grad_norm": 4.45231819152832,
+      "grad_norm": 4.019155025482178,
       "learning_rate": 0.0003904761904761905,
-      "loss": 1.4072,
+      "loss": 1.4103,
       "step": 230
     },
     {
       "epoch": 0.6857142857142857,
-      "grad_norm": 2.5132458209991455,
+      "grad_norm": 3.1268930435180664,
       "learning_rate": 0.0003857142857142857,
-      "loss": 1.2506,
+      "loss": 1.2308,
       "step": 240
     },
     {
       "epoch": 0.7142857142857143,
-      "grad_norm": 2.3154046535491943,
+      "grad_norm": 2.708326578140259,
       "learning_rate": 0.00038095238095238096,
-      "loss": 1.3371,
+      "loss": 1.3197,
       "step": 250
     },
     {
       "epoch": 0.7428571428571429,
-      "grad_norm": 2.9877989292144775,
+      "grad_norm": 2.4128565788269043,
       "learning_rate": 0.0003761904761904762,
-      "loss": 1.2249,
+      "loss": 1.1922,
       "step": 260
     },
     {
       "epoch": 0.7714285714285715,
-      "grad_norm": 4.407787799835205,
+      "grad_norm": 4.487890243530273,
       "learning_rate": 0.00037142857142857143,
-      "loss": 1.2712,
+      "loss": 1.2505,
       "step": 270
     },
     {
       "epoch": 0.8,
-      "grad_norm": 2.691910982131958,
+      "grad_norm": 3.1995606422424316,
       "learning_rate": 0.00036666666666666667,
-      "loss": 1.2366,
+      "loss": 1.2047,
       "step": 280
     },
     {
       "epoch": 0.8285714285714286,
-      "grad_norm": 2.3351523876190186,
+      "grad_norm": 3.0884156227111816,
       "learning_rate": 0.0003619047619047619,
-      "loss": 1.1777,
+      "loss": 1.1397,
       "step": 290
     },
     {
       "epoch": 0.8571428571428571,
-      "grad_norm": 2.878443956375122,
+      "grad_norm": 3.9434640407562256,
       "learning_rate": 0.00035714285714285714,
-      "loss": 1.1375,
+      "loss": 1.0844,
       "step": 300
     },
     {
       "epoch": 0.8857142857142857,
-      "grad_norm": 3.0720036029815674,
+      "grad_norm": 3.6582884788513184,
       "learning_rate": 0.00035238095238095243,
-      "loss": 1.1578,
+      "loss": 1.1461,
       "step": 310
     },
     {
       "epoch": 0.9142857142857143,
-      "grad_norm": 2.359656572341919,
+      "grad_norm": 2.6335928440093994,
       "learning_rate": 0.0003476190476190476,
-      "loss": 1.4001,
+      "loss": 1.4124,
       "step": 320
     },
     {
       "epoch": 0.9428571428571428,
-      "grad_norm": 2.3982980251312256,
+      "grad_norm": 2.8153300285339355,
       "learning_rate": 0.00034285714285714285,
-      "loss": 1.1716,
+      "loss": 1.1619,
       "step": 330
     },
     {
       "epoch": 0.9714285714285714,
-      "grad_norm": 4.372157573699951,
+      "grad_norm": 5.384872913360596,
       "learning_rate": 0.00033809523809523814,
-      "loss": 1.1177,
+      "loss": 1.1091,
       "step": 340
     },
     {
       "epoch": 1.0,
-      "grad_norm": 4.467468738555908,
+      "grad_norm": 4.4129958152771,
       "learning_rate": 0.0003333333333333333,
-      "loss": 1.1884,
+      "loss": 1.1665,
       "step": 350
     },
     {
       "epoch": 1.0,
-      "eval_loss": 1.0876593589782715,
-      "eval_runtime": 16.2651,
-      "eval_samples_per_second": 12.296,
-      "eval_steps_per_second": 1.537,
+      "eval_loss": 1.0720068216323853,
+      "eval_runtime": 16.9829,
+      "eval_samples_per_second": 11.777,
+      "eval_steps_per_second": 1.472,
       "step": 350
     },
     {
       "epoch": 1.0285714285714285,
-      "grad_norm": 8.64857292175293,
+      "grad_norm": 7.848823547363281,
       "learning_rate": 0.00032857142857142856,
-      "loss": 1.109,
+      "loss": 1.073,
       "step": 360
     },
     {
       "epoch": 1.0571428571428572,
-      "grad_norm": 3.4350078105926514,
+      "grad_norm": 3.559180498123169,
       "learning_rate": 0.00032380952380952385,
-      "loss": 1.2132,
+      "loss": 1.1773,
       "step": 370
     },
     {
       "epoch": 1.0857142857142856,
-      "grad_norm": 2.4915060997009277,
+      "grad_norm": 2.477151870727539,
       "learning_rate": 0.00031904761904761903,
-      "loss": 1.0756,
+      "loss": 1.0647,
       "step": 380
     },
     {
       "epoch": 1.1142857142857143,
-      "grad_norm": 3.537388324737549,
+      "grad_norm": 3.4121389389038086,
       "learning_rate": 0.00031428571428571427,
-      "loss": 1.0506,
+      "loss": 1.0428,
       "step": 390
     },
     {
       "epoch": 1.1428571428571428,
-      "grad_norm": 3.6561241149902344,
+      "grad_norm": 6.0573015213012695,
       "learning_rate": 0.00030952380952380956,
-      "loss": 1.1051,
+      "loss": 1.0667,
       "step": 400
     },
     {
       "epoch": 1.1714285714285715,
-      "grad_norm": 3.6229779720306396,
+      "grad_norm": 3.7485954761505127,
       "learning_rate": 0.0003047619047619048,
-      "loss": 1.0663,
+      "loss": 1.0497,
       "step": 410
     },
     {
       "epoch": 1.2,
-      "grad_norm": 4.541111946105957,
+      "grad_norm": 4.239710330963135,
       "learning_rate": 0.0003,
-      "loss": 1.1192,
+      "loss": 1.1227,
       "step": 420
     },
     {
       "epoch": 1.2285714285714286,
-      "grad_norm": 8.634881973266602,
+      "grad_norm": 7.383611679077148,
       "learning_rate": 0.00029523809523809526,
-      "loss": 1.042,
+      "loss": 1.0,
       "step": 430
     },
     {
       "epoch": 1.2571428571428571,
-      "grad_norm": 2.93104887008667,
+      "grad_norm": 3.1444430351257324,
       "learning_rate": 0.0002904761904761905,
-      "loss": 0.9288,
+      "loss": 0.9194,
       "step": 440
     },
     {
       "epoch": 1.2857142857142856,
-      "grad_norm": 6.468315601348877,
+      "grad_norm": 8.754597663879395,
       "learning_rate": 0.0002857142857142857,
-      "loss": 1.2407,
+      "loss": 1.2234,
       "step": 450
     },
     {
       "epoch": 1.3142857142857143,
-      "grad_norm": 5.183238506317139,
+      "grad_norm": 5.383545875549316,
       "learning_rate": 0.00028095238095238097,
-      "loss": 1.0302,
+      "loss": 1.0219,
       "step": 460
     },
     {
       "epoch": 1.342857142857143,
-      "grad_norm": 2.7399845123291016,
+      "grad_norm": 3.977374792098999,
       "learning_rate": 0.0002761904761904762,
-      "loss": 1.096,
+      "loss": 1.061,
       "step": 470
     },
     {
       "epoch": 1.3714285714285714,
-      "grad_norm": 4.08115291595459,
+      "grad_norm": 3.6732325553894043,
       "learning_rate": 0.0002714285714285714,
-      "loss": 1.0931,
+      "loss": 1.1131,
       "step": 480
     },
     {
       "epoch": 1.4,
-      "grad_norm": 3.567685842514038,
+      "grad_norm": 4.493168830871582,
       "learning_rate": 0.0002666666666666667,
-      "loss": 0.9909,
+      "loss": 1.0033,
       "step": 490
     },
     {
       "epoch": 1.4285714285714286,
-      "grad_norm": 3.2980871200561523,
+      "grad_norm": 2.9993937015533447,
       "learning_rate": 0.0002619047619047619,
-      "loss": 1.0007,
+      "loss": 1.0023,
       "step": 500
     },
     {
       "epoch": 1.457142857142857,
-      "grad_norm": 4.056380271911621,
+      "grad_norm": 3.984879732131958,
       "learning_rate": 0.0002571428571428571,
-      "loss": 0.9869,
+      "loss": 0.9796,
       "step": 510
     },
     {
       "epoch": 1.4857142857142858,
-      "grad_norm": 5.267114162445068,
+      "grad_norm": 5.652857303619385,
       "learning_rate": 0.0002523809523809524,
-      "loss": 0.9149,
+      "loss": 0.912,
       "step": 520
     },
     {
       "epoch": 1.5142857142857142,
-      "grad_norm": 3.0776150226593018,
+      "grad_norm": 4.275585174560547,
       "learning_rate": 0.0002476190476190476,
-      "loss": 0.9842,
+      "loss": 0.9618,
       "step": 530
     },
     {
       "epoch": 1.5428571428571427,
-      "grad_norm": 3.3917739391326904,
+      "grad_norm": 4.000895977020264,
       "learning_rate": 0.00024285714285714286,
-      "loss": 1.0165,
+      "loss": 1.0019,
       "step": 540
     },
     {
       "epoch": 1.5714285714285714,
-      "grad_norm": 3.262791633605957,
+      "grad_norm": 3.681532144546509,
       "learning_rate": 0.0002380952380952381,
-      "loss": 0.9766,
+      "loss": 0.9769,
       "step": 550
     },
     {
       "epoch": 1.6,
-      "grad_norm": 5.334137916564941,
+      "grad_norm": 6.31843376159668,
       "learning_rate": 0.00023333333333333333,
-      "loss": 1.0761,
+      "loss": 1.0508,
       "step": 560
     },
     {
       "epoch": 1.6285714285714286,
-      "grad_norm": 3.0354158878326416,
+      "grad_norm": 3.165881633758545,
       "learning_rate": 0.00022857142857142857,
-      "loss": 1.1283,
+      "loss": 1.1192,
       "step": 570
     },
     {
       "epoch": 1.657142857142857,
-      "grad_norm": 2.9675745964050293,
+      "grad_norm": 3.3488543033599854,
       "learning_rate": 0.00022380952380952383,
-      "loss": 0.9564,
+      "loss": 0.945,
       "step": 580
     },
     {
       "epoch": 1.6857142857142857,
-      "grad_norm": 3.70650053024292,
+      "grad_norm": 3.479367256164551,
       "learning_rate": 0.00021904761904761904,
-      "loss": 1.1382,
+      "loss": 1.1191,
       "step": 590
     },
     {
       "epoch": 1.7142857142857144,
-      "grad_norm": 4.117910385131836,
+      "grad_norm": 3.9680163860321045,
       "learning_rate": 0.00021428571428571427,
-      "loss": 1.0104,
+      "loss": 1.0116,
       "step": 600
     },
     {
       "epoch": 1.7428571428571429,
-      "grad_norm": 3.287292957305908,
+      "grad_norm": 3.4398531913757324,
       "learning_rate": 0.00020952380952380954,
-      "loss": 1.0152,
+      "loss": 1.0162,
       "step": 610
     },
     {
       "epoch": 1.7714285714285714,
-      "grad_norm": 4.567937850952148,
+      "grad_norm": 4.210978031158447,
       "learning_rate": 0.00020476190476190477,
-      "loss": 0.9348,
+      "loss": 0.9214,
       "step": 620
     },
     {
       "epoch": 1.8,
-      "grad_norm": 3.143092393875122,
+      "grad_norm": 3.9029576778411865,
       "learning_rate": 0.0002,
-      "loss": 1.0373,
+      "loss": 1.0135,
       "step": 630
     },
     {
       "epoch": 1.8285714285714287,
-      "grad_norm": 3.003242015838623,
+      "grad_norm": 3.1066203117370605,
       "learning_rate": 0.00019523809523809525,
-      "loss": 1.0157,
+      "loss": 1.0136,
       "step": 640
     },
     {
       "epoch": 1.8571428571428572,
-      "grad_norm": 5.4209489822387695,
+      "grad_norm": 4.775870323181152,
       "learning_rate": 0.00019047619047619048,
-      "loss": 0.8729,
+      "loss": 0.8558,
       "step": 650
     },
     {
       "epoch": 1.8857142857142857,
-      "grad_norm": 8.428689956665039,
+      "grad_norm": 4.7907490730285645,
       "learning_rate": 0.00018571428571428572,
-      "loss": 0.9703,
+      "loss": 0.9436,
       "step": 660
     },
     {
       "epoch": 1.9142857142857141,
-      "grad_norm": 4.8756866455078125,
+      "grad_norm": 4.235854148864746,
       "learning_rate": 0.00018095238095238095,
-      "loss": 1.0013,
+      "loss": 0.9884,
       "step": 670
     },
     {
       "epoch": 1.9428571428571428,
-      "grad_norm": 3.252725601196289,
+      "grad_norm": 3.239032030105591,
       "learning_rate": 0.00017619047619047622,
-      "loss": 0.9414,
+      "loss": 0.9304,
       "step": 680
     },
     {
       "epoch": 1.9714285714285715,
-      "grad_norm": 3.5052497386932373,
+      "grad_norm": 3.5313665866851807,
       "learning_rate": 0.00017142857142857143,
-      "loss": 0.9891,
+      "loss": 0.983,
       "step": 690
     },
     {
       "epoch": 2.0,
-      "grad_norm": 2.886568546295166,
+      "grad_norm": 3.0609097480773926,
       "learning_rate": 0.00016666666666666666,
-      "loss": 1.0002,
+      "loss": 0.9815,
       "step": 700
     },
     {
       "epoch": 2.0,
-      "eval_loss": 0.9278395175933838,
-      "eval_runtime": 16.2518,
-      "eval_samples_per_second": 12.306,
-      "eval_steps_per_second": 1.538,
+      "eval_loss": 0.8912971615791321,
+      "eval_runtime": 17.0248,
+      "eval_samples_per_second": 11.748,
+      "eval_steps_per_second": 1.468,
       "step": 700
     }
   ],