Training in progress, step 8000, checkpoint

last-checkpoint/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6b71114f676b03c0bcb0f46c8fd50023e59488725d453b15759a44f3e2d04275
+oid sha256:ee1d47b290561c1f0fe6c6b8eeb2bbdab68f2d3b8efe56b17627e16c8f0b3e90
 size 499730128

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fe373abbfe5cca2334d2427e9171b4bb2c3f4a319ffc60e24e707b0beb7e2fc4
-size 1442699
+oid sha256:c5c83139244a7108921d98781b19cbbcbb18cb79303c29047228167d7892c791
+size 812034507

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:15bd918144b9671ae295d46e463180bdad2e73e51a7670636feb8fc58e14c426
+oid sha256:231a1d74fdd644709c48076ea6500b1a4f8ca1620a6f210d62a4c539f1ee0b0f
 size 14645

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:76cf98a8cde877f1e49a9e4b884c4dce3d51bb4035429d464e74acc35ddca024
+oid sha256:e9567d76a2efaa946d19913fd9fe6f6f70f6f7101d7dcb8866c09743534f2d4d
 size 1465

last-checkpoint/trainer_state.json

@@ -2,328 +2,644 @@
   "best_global_step": null,
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 0.08,
+  "epoch": 0.16,
   "eval_steps": 1000,
-  "global_step": 4000,
+  "global_step": 8000,
   "is_hyper_param_search": false,
-  "is_local_process_zero": false,
+  "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
       "epoch": 0.002,
-      "grad_norm": 19.660429000854492,
+      "grad_norm": 1.2400550842285156,
       "learning_rate": 7.920000000000001e-05,
-      "loss": 67.8543,
+      "loss": 8.362694702148438,
       "step": 100
     },
     {
       "epoch": 0.004,
-      "grad_norm": 24.802082061767578,
+      "grad_norm": 2.1569149494171143,
       "learning_rate": 0.00015920000000000002,
-      "loss": 54.1444,
+      "loss": 6.592593383789063,
       "step": 200
     },
     {
       "epoch": 0.006,
-      "grad_norm": 16.059066772460938,
+      "grad_norm": 4.070979595184326,
       "learning_rate": 0.00023920000000000001,
-      "loss": 51.5921,
+      "loss": 6.208985595703125,
       "step": 300
     },
     {
       "epoch": 0.008,
-      "grad_norm": 27.705520629882812,
+      "grad_norm": 4.392622470855713,
       "learning_rate": 0.0003192,
-      "loss": 50.0536,
+      "loss": 5.921519775390625,
       "step": 400
     },
     {
       "epoch": 0.01,
-      "grad_norm": 25.79052734375,
+      "grad_norm": 4.653286457061768,
       "learning_rate": 0.0003992,
-      "loss": 48.6987,
+      "loss": 5.658856201171875,
       "step": 500
     },
     {
       "epoch": 0.012,
-      "grad_norm": 14.82126522064209,
+      "grad_norm": 2.6610119342803955,
       "learning_rate": 0.00047920000000000005,
-      "loss": 47.3182,
+      "loss": 5.450134887695312,
       "step": 600
     },
     {
       "epoch": 0.014,
-      "grad_norm": 12.00988483428955,
+      "grad_norm": 2.6298203468322754,
       "learning_rate": 0.0005592,
-      "loss": 46.6574,
+      "loss": 5.260469360351562,
       "step": 700
     },
     {
       "epoch": 0.016,
-      "grad_norm": 32.48664474487305,
+      "grad_norm": 1.681970238685608,
       "learning_rate": 0.0006392,
-      "loss": 46.0747,
+      "loss": 5.112535095214843,
       "step": 800
     },
     {
       "epoch": 0.018,
-      "grad_norm": 25.646282196044922,
+      "grad_norm": 1.5673184394836426,
       "learning_rate": 0.0007191999999999999,
-      "loss": 45.6098,
+      "loss": 5.0076907348632815,
       "step": 900
     },
     {
       "epoch": 0.02,
-      "grad_norm": 16.071203231811523,
+      "grad_norm": 1.6654455661773682,
       "learning_rate": 0.0007992,
-      "loss": 44.8537,
+      "loss": 4.9159228515625,
       "step": 1000
     },
     {
       "epoch": 0.02,
-      "eval_accuracy": 0.18606457925636008,
-      "eval_loss": 44.18140411376953,
-      "eval_runtime": 14.581,
-      "eval_samples_per_second": 8.573,
-      "eval_steps_per_second": 0.137,
+      "eval_accuracy": 0.2274637964774951,
+      "eval_loss": 4.8702616691589355,
+      "eval_runtime": 5.1375,
+      "eval_samples_per_second": 194.646,
+      "eval_steps_per_second": 1.557,
       "step": 1000
     },
     {
       "epoch": 0.022,
-      "grad_norm": 14.277446746826172,
+      "grad_norm": 1.3562606573104858,
       "learning_rate": 0.0008792,
-      "loss": 44.223,
+      "loss": 4.811459045410157,
       "step": 1100
     },
     {
       "epoch": 0.024,
-      "grad_norm": 9.809870719909668,
+      "grad_norm": 1.1230770349502563,
       "learning_rate": 0.0009592000000000001,
-      "loss": 43.5719,
+      "loss": 4.759424438476563,
       "step": 1200
     },
     {
       "epoch": 0.026,
-      "grad_norm": 16.994014739990234,
+      "grad_norm": 0.9026045203208923,
       "learning_rate": 0.0010391999999999999,
-      "loss": 43.0323,
+      "loss": 4.691336669921875,
       "step": 1300
     },
     {
       "epoch": 0.028,
-      "grad_norm": 8.082005500793457,
+      "grad_norm": 0.8914104104042053,
       "learning_rate": 0.0011192,
-      "loss": 42.3631,
+      "loss": 4.6415283203125,
       "step": 1400
     },
     {
       "epoch": 0.03,
-      "grad_norm": 7.080636024475098,
+      "grad_norm": 0.8752658367156982,
       "learning_rate": 0.0011992,
-      "loss": 42.0285,
+      "loss": 4.560531311035156,
       "step": 1500
     },
     {
       "epoch": 0.032,
-      "grad_norm": 8.547815322875977,
+      "grad_norm": 0.7435582280158997,
       "learning_rate": 0.0012791999999999999,
-      "loss": 42.6829,
+      "loss": 4.516304016113281,
       "step": 1600
     },
     {
       "epoch": 0.034,
-      "grad_norm": 7.541939735412598,
+      "grad_norm": 0.7041499018669128,
       "learning_rate": 0.0013592,
-      "loss": 41.9796,
+      "loss": 4.48115234375,
       "step": 1700
     },
     {
       "epoch": 0.036,
-      "grad_norm": 6.902923583984375,
+      "grad_norm": 0.5327283143997192,
       "learning_rate": 0.0014392,
-      "loss": 41.5016,
+      "loss": 4.429443969726562,
       "step": 1800
     },
     {
       "epoch": 0.038,
-      "grad_norm": 7.68479061126709,
+      "grad_norm": 0.45207348465919495,
       "learning_rate": 0.0015192,
-      "loss": 41.2607,
+      "loss": 4.37362060546875,
       "step": 1900
     },
     {
       "epoch": 0.04,
-      "grad_norm": 8.936702728271484,
+      "grad_norm": 0.5152547359466553,
       "learning_rate": 0.0015992,
-      "loss": 40.7545,
+      "loss": 4.377236938476562,
       "step": 2000
     },
     {
       "epoch": 0.04,
-      "eval_accuracy": 0.21657142857142858,
-      "eval_loss": 40.417686462402344,
-      "eval_runtime": 3.051,
-      "eval_samples_per_second": 40.97,
-      "eval_steps_per_second": 0.656,
+      "eval_accuracy": 0.2694227005870842,
+      "eval_loss": 4.3582682609558105,
+      "eval_runtime": 5.3878,
+      "eval_samples_per_second": 185.606,
+      "eval_steps_per_second": 1.485,
       "step": 2000
     },
     {
       "epoch": 0.042,
-      "grad_norm": 5.834764003753662,
+      "grad_norm": 0.45765218138694763,
       "learning_rate": 0.0016792,
-      "loss": 40.2024,
+      "loss": 4.341053161621094,
       "step": 2100
     },
     {
       "epoch": 0.044,
-      "grad_norm": 3.9864397048950195,
+      "grad_norm": 0.3776390254497528,
       "learning_rate": 0.0017592,
-      "loss": 39.4534,
+      "loss": 4.304255676269531,
       "step": 2200
     },
     {
       "epoch": 0.046,
-      "grad_norm": 6.052990436553955,
+      "grad_norm": 0.48194995522499084,
       "learning_rate": 0.0018392,
-      "loss": 40.0088,
+      "loss": 4.308780822753906,
       "step": 2300
     },
     {
       "epoch": 0.048,
-      "grad_norm": 10.583518981933594,
+      "grad_norm": 0.40603286027908325,
       "learning_rate": 0.0019192,
-      "loss": 40.099,
+      "loss": 4.272806396484375,
       "step": 2400
     },
     {
       "epoch": 0.05,
-      "grad_norm": 3.686554193496704,
+      "grad_norm": 0.41386568546295166,
       "learning_rate": 0.0019992,
-      "loss": 39.5664,
+      "loss": 4.267073974609375,
       "step": 2500
     },
     {
       "epoch": 0.052,
-      "grad_norm": 2.981701612472534,
+      "grad_norm": 0.4916844666004181,
       "learning_rate": 0.001999978563623903,
-      "loss": 39.5087,
+      "loss": 4.220490417480469,
       "step": 2600
     },
     {
       "epoch": 0.054,
-      "grad_norm": 3.382185459136963,
+      "grad_norm": 0.37477609515190125,
       "learning_rate": 0.0019999133871331223,
-      "loss": 38.9185,
+      "loss": 4.224794006347656,
       "step": 2700
     },
     {
       "epoch": 0.056,
-      "grad_norm": 2.8874025344848633,
+      "grad_norm": 0.405086874961853,
       "learning_rate": 0.0019998044711915177,
-      "loss": 38.1472,
+      "loss": 4.185610961914063,
       "step": 2800
     },
     {
       "epoch": 0.058,
-      "grad_norm": 3.083789825439453,
+      "grad_norm": 0.334581196308136,
       "learning_rate": 0.0019996518205634257,
-      "loss": 37.5833,
+      "loss": 4.165193176269531,
       "step": 2900
     },
     {
       "epoch": 0.06,
-      "grad_norm": 2.315279483795166,
+      "grad_norm": 0.3760662376880646,
       "learning_rate": 0.0019994554419262797,
-      "loss": 37.927,
+      "loss": 4.161107177734375,
       "step": 3000
     },
     {
       "epoch": 0.06,
-      "eval_accuracy": 0.23828180039138944,
-      "eval_loss": 37.86294174194336,
-      "eval_runtime": 3.1437,
-      "eval_samples_per_second": 39.763,
-      "eval_steps_per_second": 0.636,
+      "eval_accuracy": 0.28686888454011744,
+      "eval_loss": 4.15764856338501,
+      "eval_runtime": 5.1036,
+      "eval_samples_per_second": 195.938,
+      "eval_steps_per_second": 1.568,
       "step": 3000
     },
     {
       "epoch": 0.062,
-      "grad_norm": 4.6366777420043945,
+      "grad_norm": 0.38909581303596497,
       "learning_rate": 0.001999215343870317,
-      "loss": 38.432,
+      "loss": 4.1489105224609375,
       "step": 3100
     },
     {
       "epoch": 0.064,
-      "grad_norm": 2.8195507526397705,
+      "grad_norm": 0.4108506143093109,
       "learning_rate": 0.0019989315368982054,
-      "loss": 37.8144,
+      "loss": 4.123897705078125,
       "step": 3200
     },
     {
       "epoch": 0.066,
-      "grad_norm": 2.320726156234741,
+      "grad_norm": 0.27484413981437683,
       "learning_rate": 0.0019986040334245797,
-      "loss": 38.1739,
+      "loss": 4.101358032226562,
       "step": 3300
     },
     {
       "epoch": 0.068,
-      "grad_norm": 2.575469732284546,
+      "grad_norm": 0.29958581924438477,
       "learning_rate": 0.001998232847775504,
-      "loss": 37.1996,
+      "loss": 4.10423583984375,
       "step": 3400
     },
     {
       "epoch": 0.07,
-      "grad_norm": 2.4692764282226562,
+      "grad_norm": 0.29483968019485474,
       "learning_rate": 0.0019978179961878404,
-      "loss": 36.8725,
+      "loss": 4.078011169433593,
       "step": 3500
     },
     {
       "epoch": 0.072,
-      "grad_norm": 2.555267810821533,
+      "grad_norm": 0.36317598819732666,
       "learning_rate": 0.001997359496808541,
-      "loss": 36.2038,
+      "loss": 4.076428527832031,
       "step": 3600
     },
     {
       "epoch": 0.074,
-      "grad_norm": 1.9989144802093506,
+      "grad_norm": 0.34998658299446106,
       "learning_rate": 0.001996857369693855,
-      "loss": 36.3164,
+      "loss": 4.052371826171875,
       "step": 3700
     },
     {
       "epoch": 0.076,
-      "grad_norm": 2.223381519317627,
+      "grad_norm": 0.31070172786712646,
       "learning_rate": 0.0019963116368084486,
-      "loss": 37.1086,
+      "loss": 4.059148559570312,
       "step": 3800
     },
     {
       "epoch": 0.078,
-      "grad_norm": 1.996813416481018,
+      "grad_norm": 0.2655385732650757,
       "learning_rate": 0.001995722322024446,
-      "loss": 36.5327,
+      "loss": 4.039622802734375,
       "step": 3900
     },
     {
       "epoch": 0.08,
-      "grad_norm": 1.8378113508224487,
+      "grad_norm": 0.4650065302848816,
       "learning_rate": 0.001995089451120385,
-      "loss": 36.314,
+      "loss": 4.011789855957031,
       "step": 4000
     },
     {
       "epoch": 0.08,
-      "eval_accuracy": 0.2553561643835616,
-      "eval_loss": 36.08281707763672,
-      "eval_runtime": 3.1615,
-      "eval_samples_per_second": 39.538,
-      "eval_steps_per_second": 0.633,
+      "eval_accuracy": 0.29829745596868884,
+      "eval_loss": 4.0341691970825195,
+      "eval_runtime": 5.1511,
+      "eval_samples_per_second": 194.133,
+      "eval_steps_per_second": 1.553,
       "step": 4000
+    },
+    {
+      "epoch": 0.082,
+      "grad_norm": 0.43114492297172546,
+      "learning_rate": 0.0019944130517800893,
+      "loss": 4.0238558959960935,
+      "step": 4100
+    },
+    {
+      "epoch": 0.084,
+      "grad_norm": 0.26026225090026855,
+      "learning_rate": 0.001993693153591457,
+      "loss": 4.017567443847656,
+      "step": 4200
+    },
+    {
+      "epoch": 0.086,
+      "grad_norm": 0.23712828755378723,
+      "learning_rate": 0.0019929297880451674,
+      "loss": 3.9972265625,
+      "step": 4300
+    },
+    {
+      "epoch": 0.088,
+      "grad_norm": 0.44415482878685,
+      "learning_rate": 0.0019921229885333023,
+      "loss": 3.9874578857421876,
+      "step": 4400
+    },
+    {
+      "epoch": 0.09,
+      "grad_norm": 0.27471208572387695,
+      "learning_rate": 0.001991272790347886,
+      "loss": 3.998799133300781,
+      "step": 4500
+    },
+    {
+      "epoch": 0.092,
+      "grad_norm": 0.2248983383178711,
+      "learning_rate": 0.0019903792306793415,
+      "loss": 3.9638967895507813,
+      "step": 4600
+    },
+    {
+      "epoch": 0.094,
+      "grad_norm": 0.36029258370399475,
+      "learning_rate": 0.001989442348614863,
+      "loss": 3.972870178222656,
+      "step": 4700
+    },
+    {
+      "epoch": 0.096,
+      "grad_norm": 0.26471713185310364,
+      "learning_rate": 0.0019884621851367075,
+      "loss": 3.959658508300781,
+      "step": 4800
+    },
+    {
+      "epoch": 0.098,
+      "grad_norm": 0.23001010715961456,
+      "learning_rate": 0.001987438783120401,
+      "loss": 3.9595660400390624,
+      "step": 4900
+    },
+    {
+      "epoch": 0.1,
+      "grad_norm": 0.23091939091682434,
+      "learning_rate": 0.001986372187332862,
+      "loss": 3.9660198974609373,
+      "step": 5000
+    },
+    {
+      "epoch": 0.1,
+      "eval_accuracy": 0.3069041095890411,
+      "eval_loss": 3.95263409614563,
+      "eval_runtime": 5.269,
+      "eval_samples_per_second": 189.788,
+      "eval_steps_per_second": 1.518,
+      "step": 5000
+    },
+    {
+      "epoch": 0.102,
+      "grad_norm": 0.28329843282699585,
+      "learning_rate": 0.0019852624444304467,
+      "loss": 3.9279766845703126,
+      "step": 5100
+    },
+    {
+      "epoch": 0.104,
+      "grad_norm": 0.4390493333339691,
+      "learning_rate": 0.0019841096029569044,
+      "loss": 3.940526123046875,
+      "step": 5200
+    },
+    {
+      "epoch": 0.106,
+      "grad_norm": 0.22625097632408142,
+      "learning_rate": 0.0019829137133412556,
+      "loss": 3.9342303466796875,
+      "step": 5300
+    },
+    {
+      "epoch": 0.108,
+      "grad_norm": 0.21681281924247742,
+      "learning_rate": 0.001981674827895587,
+      "loss": 3.9172735595703125,
+      "step": 5400
+    },
+    {
+      "epoch": 0.11,
+      "grad_norm": 0.3054690957069397,
+      "learning_rate": 0.00198039300081276,
+      "loss": 3.920041198730469,
+      "step": 5500
+    },
+    {
+      "epoch": 0.112,
+      "grad_norm": 0.24492567777633667,
+      "learning_rate": 0.0019790682881640448,
+      "loss": 3.918515319824219,
+      "step": 5600
+    },
+    {
+      "epoch": 0.114,
+      "grad_norm": 0.20615942776203156,
+      "learning_rate": 0.001977700747896664,
+      "loss": 3.902435302734375,
+      "step": 5700
+    },
+    {
+      "epoch": 0.116,
+      "grad_norm": 0.2998456358909607,
+      "learning_rate": 0.001976290439831259,
+      "loss": 3.9016647338867188,
+      "step": 5800
+    },
+    {
+      "epoch": 0.118,
+      "grad_norm": 0.22004957497119904,
+      "learning_rate": 0.0019748374256592736,
+      "loss": 3.897244567871094,
+      "step": 5900
+    },
+    {
+      "epoch": 0.12,
+      "grad_norm": 0.2709659934043884,
+      "learning_rate": 0.0019733417689402543,
+      "loss": 3.892081604003906,
+      "step": 6000
+    },
+    {
+      "epoch": 0.12,
+      "eval_accuracy": 0.31208414872798435,
+      "eval_loss": 3.8932552337646484,
+      "eval_runtime": 5.3089,
+      "eval_samples_per_second": 188.362,
+      "eval_steps_per_second": 1.507,
+      "step": 6000
+    },
+    {
+      "epoch": 0.122,
+      "grad_norm": 0.38190293312072754,
+      "learning_rate": 0.0019718035350990712,
+      "loss": 3.892044982910156,
+      "step": 6100
+    },
+    {
+      "epoch": 0.124,
+      "grad_norm": 0.26727622747421265,
+      "learning_rate": 0.0019702227914230566,
+      "loss": 3.8680352783203125,
+      "step": 6200
+    },
+    {
+      "epoch": 0.126,
+      "grad_norm": 0.315392404794693,
+      "learning_rate": 0.001968599607059059,
+      "loss": 3.8766015625,
+      "step": 6300
+    },
+    {
+      "epoch": 0.128,
+      "grad_norm": 0.25175315141677856,
+      "learning_rate": 0.0019669340530104207,
+      "loss": 3.8739111328125,
+      "step": 6400
+    },
+    {
+      "epoch": 0.13,
+      "grad_norm": 0.19557097554206848,
+      "learning_rate": 0.001965226202133872,
+      "loss": 3.8594989013671874,
+      "step": 6500
+    },
+    {
+      "epoch": 0.132,
+      "grad_norm": 0.22091643512248993,
+      "learning_rate": 0.0019634761291363427,
+      "loss": 3.8679965209960936,
+      "step": 6600
+    },
+    {
+      "epoch": 0.134,
+      "grad_norm": 0.2809944152832031,
+      "learning_rate": 0.0019616839105716954,
+      "loss": 3.8709503173828126,
+      "step": 6700
+    },
+    {
+      "epoch": 0.136,
+      "grad_norm": 0.24812288582324982,
+      "learning_rate": 0.0019598496248373755,
+      "loss": 3.834676513671875,
+      "step": 6800
+    },
+    {
+      "epoch": 0.138,
+      "grad_norm": 0.218924418091774,
+      "learning_rate": 0.001957973352170984,
+      "loss": 3.8453244018554686,
+      "step": 6900
+    },
+    {
+      "epoch": 0.14,
+      "grad_norm": 0.3137245774269104,
+      "learning_rate": 0.001956055174646765,
+      "loss": 3.841236572265625,
+      "step": 7000
+    },
+    {
+      "epoch": 0.14,
+      "eval_accuracy": 0.3165616438356164,
+      "eval_loss": 3.850874900817871,
+      "eval_runtime": 5.2575,
+      "eval_samples_per_second": 190.203,
+      "eval_steps_per_second": 1.522,
+      "step": 7000
+    },
+    {
+      "epoch": 0.142,
+      "grad_norm": 0.22134174406528473,
+      "learning_rate": 0.0019540951761720174,
+      "loss": 3.842283020019531,
+      "step": 7100
+    },
+    {
+      "epoch": 0.144,
+      "grad_norm": 0.27472418546676636,
+      "learning_rate": 0.0019520934424834247,
+      "loss": 3.8458221435546873,
+      "step": 7200
+    },
+    {
+      "epoch": 0.146,
+      "grad_norm": 0.23996180295944214,
+      "learning_rate": 0.0019500500611433025,
+      "loss": 3.817427673339844,
+      "step": 7300
+    },
+    {
+      "epoch": 0.148,
+      "grad_norm": 0.2116343080997467,
+      "learning_rate": 0.0019479651215357707,
+      "loss": 3.831176452636719,
+      "step": 7400
+    },
+    {
+      "epoch": 0.15,
+      "grad_norm": 0.171181321144104,
+      "learning_rate": 0.0019458387148628417,
+      "loss": 3.832611389160156,
+      "step": 7500
+    },
+    {
+      "epoch": 0.152,
+      "grad_norm": 0.24767358601093292,
+      "learning_rate": 0.001943670934140432,
+      "loss": 3.8011544799804686,
+      "step": 7600
+    },
+    {
+      "epoch": 0.154,
+      "grad_norm": 0.31432944536209106,
+      "learning_rate": 0.0019414618741942936,
+      "loss": 3.8120742797851563,
+      "step": 7700
+    },
+    {
+      "epoch": 0.156,
+      "grad_norm": 0.20955601334571838,
+      "learning_rate": 0.0019392116316558638,
+      "loss": 3.8340234375,
+      "step": 7800
+    },
+    {
+      "epoch": 0.158,
+      "grad_norm": 0.2197394073009491,
+      "learning_rate": 0.001936920304958042,
+      "loss": 3.8019769287109373,
+      "step": 7900
+    },
+    {
+      "epoch": 0.16,
+      "grad_norm": 0.19528572261333466,
+      "learning_rate": 0.0019345879943308804,
+      "loss": 3.802054138183594,
+      "step": 8000
+    },
+    {
+      "epoch": 0.16,
+      "eval_accuracy": 0.32047162426614484,
+      "eval_loss": 3.8114547729492188,
+      "eval_runtime": 5.3173,
+      "eval_samples_per_second": 188.067,
+      "eval_steps_per_second": 1.505,
+      "step": 8000
     }
   ],
   "logging_steps": 100,
@@ -344,7 +660,7 @@
     }
   },
   "total_flos": 0.0,
-  "train_batch_size": 8,
+  "train_batch_size": 128,
   "trial_name": null,
   "trial_params": null
 }

last-checkpoint/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:360066b4a5f1cc8c89941d389baaf9061071c4da2a624ff0fb9b3a9211c62ff7
-size 5905
+oid sha256:3ab34ea13b0e64af4c31a655731e444610b50d6a753b685ccf0d645cea5f76a3
+size 5265