End of training

README.md

@@ -21,7 +21,7 @@ model-index:
     metrics:
     - name: Accuracy
       type: accuracy
-      value: 0.7608695652173914
+      value: 0.8043478260869565
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -31,8 +31,8 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [MBZUAI/swiftformer-xs](https://huggingface.co/MBZUAI/swiftformer-xs) on the imagefolder dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.8869
-- Accuracy: 0.7609
+- Loss: 0.7032
+- Accuracy: 0.8043
 
 ## Model description
 

all_results.json

@@ -0,0 +1,12 @@
+{
+    "epoch": 73.85,
+    "eval_accuracy": 0.8043478260869565,
+    "eval_loss": 0.7032170295715332,
+    "eval_runtime": 0.8089,
+    "eval_samples_per_second": 56.865,
+    "eval_steps_per_second": 2.472,
+    "train_loss": 0.40781121912101903,
+    "train_runtime": 277.6094,
+    "train_samples_per_second": 236.015,
+    "train_steps_per_second": 1.729
+}

eval_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 73.85,
+    "eval_accuracy": 0.8043478260869565,
+    "eval_loss": 0.7032170295715332,
+    "eval_runtime": 0.8089,
+    "eval_samples_per_second": 56.865,
+    "eval_steps_per_second": 2.472
+}

runs/Jun12_13-24-49_DESKTOP-SKBE9FB/events.out.tfevents.1718220574.DESKTOP-SKBE9FB.8880.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7537522c268e66a9feb2886c2413b5eb937c2982fa7388a65a843b7ce4699729
+size 411

train_results.json

@@ -0,0 +1,7 @@
+{
+    "epoch": 73.85,
+    "train_loss": 0.40781121912101903,
+    "train_runtime": 277.6094,
+    "train_samples_per_second": 236.015,
+    "train_steps_per_second": 1.729
+}

trainer_state.json

@@ -0,0 +1,984 @@
+{
+  "best_metric": 0.8043478260869565,
+  "best_model_checkpoint": "swiftformer-xs-ve-U13-b-80c\\checkpoint-175",
+  "epoch": 73.84615384615384,
+  "eval_steps": 500,
+  "global_step": 480,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.92,
+      "eval_accuracy": 0.2391304347826087,
+      "eval_loss": 1.3860037326812744,
+      "eval_runtime": 0.6497,
+      "eval_samples_per_second": 70.799,
+      "eval_steps_per_second": 3.078,
+      "step": 6
+    },
+    {
+      "epoch": 1.54,
+      "learning_rate": 2.777777777777778e-05,
+      "loss": 1.3859,
+      "step": 10
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.30434782608695654,
+      "eval_loss": 1.3843744993209839,
+      "eval_runtime": 0.6508,
+      "eval_samples_per_second": 70.684,
+      "eval_steps_per_second": 3.073,
+      "step": 13
+    },
+    {
+      "epoch": 2.92,
+      "eval_accuracy": 0.1956521739130435,
+      "eval_loss": 1.3819622993469238,
+      "eval_runtime": 0.6238,
+      "eval_samples_per_second": 73.745,
+      "eval_steps_per_second": 3.206,
+      "step": 19
+    },
+    {
+      "epoch": 3.08,
+      "learning_rate": 5.555555555555556e-05,
+      "loss": 1.381,
+      "step": 20
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.17391304347826086,
+      "eval_loss": 1.3745858669281006,
+      "eval_runtime": 0.6174,
+      "eval_samples_per_second": 74.504,
+      "eval_steps_per_second": 3.239,
+      "step": 26
+    },
+    {
+      "epoch": 4.62,
+      "learning_rate": 8.333333333333334e-05,
+      "loss": 1.3573,
+      "step": 30
+    },
+    {
+      "epoch": 4.92,
+      "eval_accuracy": 0.1956521739130435,
+      "eval_loss": 1.3642687797546387,
+      "eval_runtime": 0.5993,
+      "eval_samples_per_second": 76.759,
+      "eval_steps_per_second": 3.337,
+      "step": 32
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.15217391304347827,
+      "eval_loss": 1.3561071157455444,
+      "eval_runtime": 0.6092,
+      "eval_samples_per_second": 75.505,
+      "eval_steps_per_second": 3.283,
+      "step": 39
+    },
+    {
+      "epoch": 6.15,
+      "learning_rate": 0.00011111111111111112,
+      "loss": 1.2692,
+      "step": 40
+    },
+    {
+      "epoch": 6.92,
+      "eval_accuracy": 0.15217391304347827,
+      "eval_loss": 1.3582805395126343,
+      "eval_runtime": 0.6066,
+      "eval_samples_per_second": 75.83,
+      "eval_steps_per_second": 3.297,
+      "step": 45
+    },
+    {
+      "epoch": 7.69,
+      "learning_rate": 0.0001388888888888889,
+      "loss": 1.1682,
+      "step": 50
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.17391304347826086,
+      "eval_loss": 1.3623026609420776,
+      "eval_runtime": 0.6189,
+      "eval_samples_per_second": 74.33,
+      "eval_steps_per_second": 3.232,
+      "step": 52
+    },
+    {
+      "epoch": 8.92,
+      "eval_accuracy": 0.2608695652173913,
+      "eval_loss": 1.3296467065811157,
+      "eval_runtime": 0.6038,
+      "eval_samples_per_second": 76.178,
+      "eval_steps_per_second": 3.312,
+      "step": 58
+    },
+    {
+      "epoch": 9.23,
+      "learning_rate": 0.0001666666666666667,
+      "loss": 1.1005,
+      "step": 60
+    },
+    {
+      "epoch": 10.0,
+      "eval_accuracy": 0.391304347826087,
+      "eval_loss": 1.266342282295227,
+      "eval_runtime": 0.6296,
+      "eval_samples_per_second": 73.06,
+      "eval_steps_per_second": 3.177,
+      "step": 65
+    },
+    {
+      "epoch": 10.77,
+      "learning_rate": 0.00019444444444444446,
+      "loss": 0.9884,
+      "step": 70
+    },
+    {
+      "epoch": 10.92,
+      "eval_accuracy": 0.3695652173913043,
+      "eval_loss": 1.3159973621368408,
+      "eval_runtime": 0.6014,
+      "eval_samples_per_second": 76.491,
+      "eval_steps_per_second": 3.326,
+      "step": 71
+    },
+    {
+      "epoch": 12.0,
+      "eval_accuracy": 0.4782608695652174,
+      "eval_loss": 1.1806195974349976,
+      "eval_runtime": 0.6259,
+      "eval_samples_per_second": 73.491,
+      "eval_steps_per_second": 3.195,
+      "step": 78
+    },
+    {
+      "epoch": 12.31,
+      "learning_rate": 0.000196078431372549,
+      "loss": 0.9111,
+      "step": 80
+    },
+    {
+      "epoch": 12.92,
+      "eval_accuracy": 0.6086956521739131,
+      "eval_loss": 1.15597403049469,
+      "eval_runtime": 0.6243,
+      "eval_samples_per_second": 73.68,
+      "eval_steps_per_second": 3.203,
+      "step": 84
+    },
+    {
+      "epoch": 13.85,
+      "learning_rate": 0.0001911764705882353,
+      "loss": 0.8464,
+      "step": 90
+    },
+    {
+      "epoch": 14.0,
+      "eval_accuracy": 0.5869565217391305,
+      "eval_loss": 1.1349503993988037,
+      "eval_runtime": 0.6226,
+      "eval_samples_per_second": 73.889,
+      "eval_steps_per_second": 3.213,
+      "step": 91
+    },
+    {
+      "epoch": 14.92,
+      "eval_accuracy": 0.6304347826086957,
+      "eval_loss": 1.0768355131149292,
+      "eval_runtime": 0.6092,
+      "eval_samples_per_second": 75.507,
+      "eval_steps_per_second": 3.283,
+      "step": 97
+    },
+    {
+      "epoch": 15.38,
+      "learning_rate": 0.00018627450980392157,
+      "loss": 0.7768,
+      "step": 100
+    },
+    {
+      "epoch": 16.0,
+      "eval_accuracy": 0.6086956521739131,
+      "eval_loss": 0.9706609845161438,
+      "eval_runtime": 0.6399,
+      "eval_samples_per_second": 71.887,
+      "eval_steps_per_second": 3.126,
+      "step": 104
+    },
+    {
+      "epoch": 16.92,
+      "learning_rate": 0.00018137254901960786,
+      "loss": 0.6754,
+      "step": 110
+    },
+    {
+      "epoch": 16.92,
+      "eval_accuracy": 0.6521739130434783,
+      "eval_loss": 0.9544328451156616,
+      "eval_runtime": 0.5992,
+      "eval_samples_per_second": 76.775,
+      "eval_steps_per_second": 3.338,
+      "step": 110
+    },
+    {
+      "epoch": 18.0,
+      "eval_accuracy": 0.6739130434782609,
+      "eval_loss": 0.9885360598564148,
+      "eval_runtime": 0.6261,
+      "eval_samples_per_second": 73.474,
+      "eval_steps_per_second": 3.195,
+      "step": 117
+    },
+    {
+      "epoch": 18.46,
+      "learning_rate": 0.00017647058823529413,
+      "loss": 0.657,
+      "step": 120
+    },
+    {
+      "epoch": 18.92,
+      "eval_accuracy": 0.6956521739130435,
+      "eval_loss": 0.85777348279953,
+      "eval_runtime": 0.6125,
+      "eval_samples_per_second": 75.103,
+      "eval_steps_per_second": 3.265,
+      "step": 123
+    },
+    {
+      "epoch": 20.0,
+      "learning_rate": 0.0001715686274509804,
+      "loss": 0.5408,
+      "step": 130
+    },
+    {
+      "epoch": 20.0,
+      "eval_accuracy": 0.7391304347826086,
+      "eval_loss": 0.7794063687324524,
+      "eval_runtime": 0.5998,
+      "eval_samples_per_second": 76.697,
+      "eval_steps_per_second": 3.335,
+      "step": 130
+    },
+    {
+      "epoch": 20.92,
+      "eval_accuracy": 0.7391304347826086,
+      "eval_loss": 0.8072043657302856,
+      "eval_runtime": 0.6119,
+      "eval_samples_per_second": 75.172,
+      "eval_steps_per_second": 3.268,
+      "step": 136
+    },
+    {
+      "epoch": 21.54,
+      "learning_rate": 0.0001666666666666667,
+      "loss": 0.5094,
+      "step": 140
+    },
+    {
+      "epoch": 22.0,
+      "eval_accuracy": 0.6739130434782609,
+      "eval_loss": 0.7917114496231079,
+      "eval_runtime": 0.6088,
+      "eval_samples_per_second": 75.563,
+      "eval_steps_per_second": 3.285,
+      "step": 143
+    },
+    {
+      "epoch": 22.92,
+      "eval_accuracy": 0.6739130434782609,
+      "eval_loss": 0.7974965572357178,
+      "eval_runtime": 0.6475,
+      "eval_samples_per_second": 71.039,
+      "eval_steps_per_second": 3.089,
+      "step": 149
+    },
+    {
+      "epoch": 23.08,
+      "learning_rate": 0.00016176470588235295,
+      "loss": 0.4546,
+      "step": 150
+    },
+    {
+      "epoch": 24.0,
+      "eval_accuracy": 0.7608695652173914,
+      "eval_loss": 0.758342981338501,
+      "eval_runtime": 0.6038,
+      "eval_samples_per_second": 76.184,
+      "eval_steps_per_second": 3.312,
+      "step": 156
+    },
+    {
+      "epoch": 24.62,
+      "learning_rate": 0.00015686274509803922,
+      "loss": 0.3722,
+      "step": 160
+    },
+    {
+      "epoch": 24.92,
+      "eval_accuracy": 0.782608695652174,
+      "eval_loss": 0.7073833346366882,
+      "eval_runtime": 0.6225,
+      "eval_samples_per_second": 73.896,
+      "eval_steps_per_second": 3.213,
+      "step": 162
+    },
+    {
+      "epoch": 26.0,
+      "eval_accuracy": 0.7391304347826086,
+      "eval_loss": 0.6909225583076477,
+      "eval_runtime": 0.6068,
+      "eval_samples_per_second": 75.807,
+      "eval_steps_per_second": 3.296,
+      "step": 169
+    },
+    {
+      "epoch": 26.15,
+      "learning_rate": 0.00015196078431372549,
+      "loss": 0.3494,
+      "step": 170
+    },
+    {
+      "epoch": 26.92,
+      "eval_accuracy": 0.8043478260869565,
+      "eval_loss": 0.7032170295715332,
+      "eval_runtime": 0.6101,
+      "eval_samples_per_second": 75.403,
+      "eval_steps_per_second": 3.278,
+      "step": 175
+    },
+    {
+      "epoch": 27.69,
+      "learning_rate": 0.00014705882352941178,
+      "loss": 0.3092,
+      "step": 180
+    },
+    {
+      "epoch": 28.0,
+      "eval_accuracy": 0.782608695652174,
+      "eval_loss": 0.8148682713508606,
+      "eval_runtime": 0.6369,
+      "eval_samples_per_second": 72.229,
+      "eval_steps_per_second": 3.14,
+      "step": 182
+    },
+    {
+      "epoch": 28.92,
+      "eval_accuracy": 0.782608695652174,
+      "eval_loss": 0.7898127436637878,
+      "eval_runtime": 0.6193,
+      "eval_samples_per_second": 74.278,
+      "eval_steps_per_second": 3.229,
+      "step": 188
+    },
+    {
+      "epoch": 29.23,
+      "learning_rate": 0.00014215686274509804,
+      "loss": 0.2643,
+      "step": 190
+    },
+    {
+      "epoch": 30.0,
+      "eval_accuracy": 0.8043478260869565,
+      "eval_loss": 0.7311907410621643,
+      "eval_runtime": 0.6043,
+      "eval_samples_per_second": 76.119,
+      "eval_steps_per_second": 3.31,
+      "step": 195
+    },
+    {
+      "epoch": 30.77,
+      "learning_rate": 0.0001372549019607843,
+      "loss": 0.2659,
+      "step": 200
+    },
+    {
+      "epoch": 30.92,
+      "eval_accuracy": 0.717391304347826,
+      "eval_loss": 0.7597643733024597,
+      "eval_runtime": 0.6226,
+      "eval_samples_per_second": 73.885,
+      "eval_steps_per_second": 3.212,
+      "step": 201
+    },
+    {
+      "epoch": 32.0,
+      "eval_accuracy": 0.7608695652173914,
+      "eval_loss": 0.7530593276023865,
+      "eval_runtime": 0.5971,
+      "eval_samples_per_second": 77.044,
+      "eval_steps_per_second": 3.35,
+      "step": 208
+    },
+    {
+      "epoch": 32.31,
+      "learning_rate": 0.0001323529411764706,
+      "loss": 0.2298,
+      "step": 210
+    },
+    {
+      "epoch": 32.92,
+      "eval_accuracy": 0.8043478260869565,
+      "eval_loss": 0.6876691579818726,
+      "eval_runtime": 0.6354,
+      "eval_samples_per_second": 72.394,
+      "eval_steps_per_second": 3.148,
+      "step": 214
+    },
+    {
+      "epoch": 33.85,
+      "learning_rate": 0.00012745098039215687,
+      "loss": 0.2147,
+      "step": 220
+    },
+    {
+      "epoch": 34.0,
+      "eval_accuracy": 0.8043478260869565,
+      "eval_loss": 0.6864094734191895,
+      "eval_runtime": 0.6204,
+      "eval_samples_per_second": 74.15,
+      "eval_steps_per_second": 3.224,
+      "step": 221
+    },
+    {
+      "epoch": 34.92,
+      "eval_accuracy": 0.7391304347826086,
+      "eval_loss": 0.7655704617500305,
+      "eval_runtime": 0.623,
+      "eval_samples_per_second": 73.835,
+      "eval_steps_per_second": 3.21,
+      "step": 227
+    },
+    {
+      "epoch": 35.38,
+      "learning_rate": 0.00012254901960784316,
+      "loss": 0.2457,
+      "step": 230
+    },
+    {
+      "epoch": 36.0,
+      "eval_accuracy": 0.7391304347826086,
+      "eval_loss": 0.8493983149528503,
+      "eval_runtime": 0.6074,
+      "eval_samples_per_second": 75.734,
+      "eval_steps_per_second": 3.293,
+      "step": 234
+    },
+    {
+      "epoch": 36.92,
+      "learning_rate": 0.00011764705882352942,
+      "loss": 0.1905,
+      "step": 240
+    },
+    {
+      "epoch": 36.92,
+      "eval_accuracy": 0.7608695652173914,
+      "eval_loss": 0.7319269776344299,
+      "eval_runtime": 0.6155,
+      "eval_samples_per_second": 74.733,
+      "eval_steps_per_second": 3.249,
+      "step": 240
+    },
+    {
+      "epoch": 38.0,
+      "eval_accuracy": 0.6956521739130435,
+      "eval_loss": 0.8289902806282043,
+      "eval_runtime": 0.6141,
+      "eval_samples_per_second": 74.91,
+      "eval_steps_per_second": 3.257,
+      "step": 247
+    },
+    {
+      "epoch": 38.46,
+      "learning_rate": 0.0001127450980392157,
+      "loss": 0.2073,
+      "step": 250
+    },
+    {
+      "epoch": 38.92,
+      "eval_accuracy": 0.7608695652173914,
+      "eval_loss": 0.7963315844535828,
+      "eval_runtime": 0.6558,
+      "eval_samples_per_second": 70.149,
+      "eval_steps_per_second": 3.05,
+      "step": 253
+    },
+    {
+      "epoch": 40.0,
+      "learning_rate": 0.00010784313725490196,
+      "loss": 0.1603,
+      "step": 260
+    },
+    {
+      "epoch": 40.0,
+      "eval_accuracy": 0.6956521739130435,
+      "eval_loss": 0.8692622780799866,
+      "eval_runtime": 0.5964,
+      "eval_samples_per_second": 77.127,
+      "eval_steps_per_second": 3.353,
+      "step": 260
+    },
+    {
+      "epoch": 40.92,
+      "eval_accuracy": 0.8043478260869565,
+      "eval_loss": 0.7137989401817322,
+      "eval_runtime": 0.6168,
+      "eval_samples_per_second": 74.577,
+      "eval_steps_per_second": 3.242,
+      "step": 266
+    },
+    {
+      "epoch": 41.54,
+      "learning_rate": 0.00010294117647058823,
+      "loss": 0.1852,
+      "step": 270
+    },
+    {
+      "epoch": 42.0,
+      "eval_accuracy": 0.7608695652173914,
+      "eval_loss": 0.7274174690246582,
+      "eval_runtime": 0.6771,
+      "eval_samples_per_second": 67.936,
+      "eval_steps_per_second": 2.954,
+      "step": 273
+    },
+    {
+      "epoch": 42.92,
+      "eval_accuracy": 0.6739130434782609,
+      "eval_loss": 0.8352705240249634,
+      "eval_runtime": 0.6424,
+      "eval_samples_per_second": 71.602,
+      "eval_steps_per_second": 3.113,
+      "step": 279
+    },
+    {
+      "epoch": 43.08,
+      "learning_rate": 9.80392156862745e-05,
+      "loss": 0.1641,
+      "step": 280
+    },
+    {
+      "epoch": 44.0,
+      "eval_accuracy": 0.6956521739130435,
+      "eval_loss": 0.9382135272026062,
+      "eval_runtime": 0.6048,
+      "eval_samples_per_second": 76.054,
+      "eval_steps_per_second": 3.307,
+      "step": 286
+    },
+    {
+      "epoch": 44.62,
+      "learning_rate": 9.313725490196079e-05,
+      "loss": 0.1568,
+      "step": 290
+    },
+    {
+      "epoch": 44.92,
+      "eval_accuracy": 0.717391304347826,
+      "eval_loss": 0.8655094504356384,
+      "eval_runtime": 0.6326,
+      "eval_samples_per_second": 72.716,
+      "eval_steps_per_second": 3.162,
+      "step": 292
+    },
+    {
+      "epoch": 46.0,
+      "eval_accuracy": 0.7391304347826086,
+      "eval_loss": 0.7620847821235657,
+      "eval_runtime": 0.6048,
+      "eval_samples_per_second": 76.056,
+      "eval_steps_per_second": 3.307,
+      "step": 299
+    },
+    {
+      "epoch": 46.15,
+      "learning_rate": 8.823529411764706e-05,
+      "loss": 0.1498,
+      "step": 300
+    },
+    {
+      "epoch": 46.92,
+      "eval_accuracy": 0.717391304347826,
+      "eval_loss": 0.794407069683075,
+      "eval_runtime": 0.6439,
+      "eval_samples_per_second": 71.44,
+      "eval_steps_per_second": 3.106,
+      "step": 305
+    },
+    {
+      "epoch": 47.69,
+      "learning_rate": 8.333333333333334e-05,
+      "loss": 0.1563,
+      "step": 310
+    },
+    {
+      "epoch": 48.0,
+      "eval_accuracy": 0.6956521739130435,
+      "eval_loss": 0.8432682752609253,
+      "eval_runtime": 0.6319,
+      "eval_samples_per_second": 72.801,
+      "eval_steps_per_second": 3.165,
+      "step": 312
+    },
+    {
+      "epoch": 48.92,
+      "eval_accuracy": 0.7608695652173914,
+      "eval_loss": 0.8633100986480713,
+      "eval_runtime": 0.6452,
+      "eval_samples_per_second": 71.293,
+      "eval_steps_per_second": 3.1,
+      "step": 318
+    },
+    {
+      "epoch": 49.23,
+      "learning_rate": 7.843137254901961e-05,
+      "loss": 0.1554,
+      "step": 320
+    },
+    {
+      "epoch": 50.0,
+      "eval_accuracy": 0.7391304347826086,
+      "eval_loss": 0.8542913198471069,
+      "eval_runtime": 0.6393,
+      "eval_samples_per_second": 71.95,
+      "eval_steps_per_second": 3.128,
+      "step": 325
+    },
+    {
+      "epoch": 50.77,
+      "learning_rate": 7.352941176470589e-05,
+      "loss": 0.1316,
+      "step": 330
+    },
+    {
+      "epoch": 50.92,
+      "eval_accuracy": 0.717391304347826,
+      "eval_loss": 0.9127072095870972,
+      "eval_runtime": 0.6018,
+      "eval_samples_per_second": 76.431,
+      "eval_steps_per_second": 3.323,
+      "step": 331
+    },
+    {
+      "epoch": 52.0,
+      "eval_accuracy": 0.6956521739130435,
+      "eval_loss": 0.9248411059379578,
+      "eval_runtime": 0.6005,
+      "eval_samples_per_second": 76.606,
+      "eval_steps_per_second": 3.331,
+      "step": 338
+    },
+    {
+      "epoch": 52.31,
+      "learning_rate": 6.862745098039216e-05,
+      "loss": 0.1264,
+      "step": 340
+    },
+    {
+      "epoch": 52.92,
+      "eval_accuracy": 0.6956521739130435,
+      "eval_loss": 0.9349088072776794,
+      "eval_runtime": 0.5946,
+      "eval_samples_per_second": 77.368,
+      "eval_steps_per_second": 3.364,
+      "step": 344
+    },
+    {
+      "epoch": 53.85,
+      "learning_rate": 6.372549019607843e-05,
+      "loss": 0.1082,
+      "step": 350
+    },
+    {
+      "epoch": 54.0,
+      "eval_accuracy": 0.6739130434782609,
+      "eval_loss": 0.978458046913147,
+      "eval_runtime": 0.6614,
+      "eval_samples_per_second": 69.548,
+      "eval_steps_per_second": 3.024,
+      "step": 351
+    },
+    {
+      "epoch": 54.92,
+      "eval_accuracy": 0.6739130434782609,
+      "eval_loss": 1.0165393352508545,
+      "eval_runtime": 0.6054,
+      "eval_samples_per_second": 75.984,
+      "eval_steps_per_second": 3.304,
+      "step": 357
+    },
+    {
+      "epoch": 55.38,
+      "learning_rate": 5.882352941176471e-05,
+      "loss": 0.1366,
+      "step": 360
+    },
+    {
+      "epoch": 56.0,
+      "eval_accuracy": 0.6956521739130435,
+      "eval_loss": 0.8369129300117493,
+      "eval_runtime": 0.6396,
+      "eval_samples_per_second": 71.916,
+      "eval_steps_per_second": 3.127,
+      "step": 364
+    },
+    {
+      "epoch": 56.92,
+      "learning_rate": 5.392156862745098e-05,
+      "loss": 0.1546,
+      "step": 370
+    },
+    {
+      "epoch": 56.92,
+      "eval_accuracy": 0.717391304347826,
+      "eval_loss": 0.8372054696083069,
+      "eval_runtime": 0.6015,
+      "eval_samples_per_second": 76.478,
+      "eval_steps_per_second": 3.325,
+      "step": 370
+    },
+    {
+      "epoch": 58.0,
+      "eval_accuracy": 0.6956521739130435,
+      "eval_loss": 0.8595665097236633,
+      "eval_runtime": 0.6434,
+      "eval_samples_per_second": 71.491,
+      "eval_steps_per_second": 3.108,
+      "step": 377
+    },
+    {
+      "epoch": 58.46,
+      "learning_rate": 4.901960784313725e-05,
+      "loss": 0.1218,
+      "step": 380
+    },
+    {
+      "epoch": 58.92,
+      "eval_accuracy": 0.717391304347826,
+      "eval_loss": 0.8054145574569702,
+      "eval_runtime": 0.6049,
+      "eval_samples_per_second": 76.042,
+      "eval_steps_per_second": 3.306,
+      "step": 383
+    },
+    {
+      "epoch": 60.0,
+      "learning_rate": 4.411764705882353e-05,
+      "loss": 0.1162,
+      "step": 390
+    },
+    {
+      "epoch": 60.0,
+      "eval_accuracy": 0.7391304347826086,
+      "eval_loss": 0.7963301539421082,
+      "eval_runtime": 0.6236,
+      "eval_samples_per_second": 73.763,
+      "eval_steps_per_second": 3.207,
+      "step": 390
+    },
+    {
+      "epoch": 60.92,
+      "eval_accuracy": 0.7391304347826086,
+      "eval_loss": 0.795264720916748,
+      "eval_runtime": 0.6138,
+      "eval_samples_per_second": 74.945,
+      "eval_steps_per_second": 3.258,
+      "step": 396
+    },
+    {
+      "epoch": 61.54,
+      "learning_rate": 3.9215686274509805e-05,
+      "loss": 0.0876,
+      "step": 400
+    },
+    {
+      "epoch": 62.0,
+      "eval_accuracy": 0.7391304347826086,
+      "eval_loss": 0.8229031562805176,
+      "eval_runtime": 0.6005,
+      "eval_samples_per_second": 76.603,
+      "eval_steps_per_second": 3.331,
+      "step": 403
+    },
+    {
+      "epoch": 62.92,
+      "eval_accuracy": 0.7391304347826086,
+      "eval_loss": 0.8364757895469666,
+      "eval_runtime": 0.626,
+      "eval_samples_per_second": 73.487,
+      "eval_steps_per_second": 3.195,
+      "step": 409
+    },
+    {
+      "epoch": 63.08,
+      "learning_rate": 3.431372549019608e-05,
+      "loss": 0.1032,
+      "step": 410
+    },
+    {
+      "epoch": 64.0,
+      "eval_accuracy": 0.7608695652173914,
+      "eval_loss": 0.8161815404891968,
+      "eval_runtime": 0.6165,
+      "eval_samples_per_second": 74.611,
+      "eval_steps_per_second": 3.244,
+      "step": 416
+    },
+    {
+      "epoch": 64.62,
+      "learning_rate": 2.9411764705882354e-05,
+      "loss": 0.0825,
+      "step": 420
+    },
+    {
+      "epoch": 64.92,
+      "eval_accuracy": 0.7391304347826086,
+      "eval_loss": 0.8645689487457275,
+      "eval_runtime": 0.608,
+      "eval_samples_per_second": 75.66,
+      "eval_steps_per_second": 3.29,
+      "step": 422
+    },
+    {
+      "epoch": 66.0,
+      "eval_accuracy": 0.7391304347826086,
+      "eval_loss": 0.9134983420372009,
+      "eval_runtime": 0.7022,
+      "eval_samples_per_second": 65.511,
+      "eval_steps_per_second": 2.848,
+      "step": 429
+    },
+    {
+      "epoch": 66.15,
+      "learning_rate": 2.4509803921568626e-05,
+      "loss": 0.1119,
+      "step": 430
+    },
+    {
+      "epoch": 66.92,
+      "eval_accuracy": 0.7391304347826086,
+      "eval_loss": 0.9164416193962097,
+      "eval_runtime": 0.6197,
+      "eval_samples_per_second": 74.228,
+      "eval_steps_per_second": 3.227,
+      "step": 435
+    },
+    {
+      "epoch": 67.69,
+      "learning_rate": 1.9607843137254903e-05,
+      "loss": 0.0949,
+      "step": 440
+    },
+    {
+      "epoch": 68.0,
+      "eval_accuracy": 0.7391304347826086,
+      "eval_loss": 0.9232246279716492,
+      "eval_runtime": 0.6325,
+      "eval_samples_per_second": 72.724,
+      "eval_steps_per_second": 3.162,
+      "step": 442
+    },
+    {
+      "epoch": 68.92,
+      "eval_accuracy": 0.7391304347826086,
+      "eval_loss": 0.9380779266357422,
+      "eval_runtime": 0.5791,
+      "eval_samples_per_second": 79.439,
+      "eval_steps_per_second": 3.454,
+      "step": 448
+    },
+    {
+      "epoch": 69.23,
+      "learning_rate": 1.4705882352941177e-05,
+      "loss": 0.1227,
+      "step": 450
+    },
+    {
+      "epoch": 70.0,
+      "eval_accuracy": 0.7391304347826086,
+      "eval_loss": 0.8997882604598999,
+      "eval_runtime": 0.606,
+      "eval_samples_per_second": 75.907,
+      "eval_steps_per_second": 3.3,
+      "step": 455
+    },
+    {
+      "epoch": 70.77,
+      "learning_rate": 9.803921568627451e-06,
+      "loss": 0.0872,
+      "step": 460
+    },
+    {
+      "epoch": 70.92,
+      "eval_accuracy": 0.717391304347826,
+      "eval_loss": 0.9632061123847961,
+      "eval_runtime": 0.5942,
+      "eval_samples_per_second": 77.419,
+      "eval_steps_per_second": 3.366,
+      "step": 461
+    },
+    {
+      "epoch": 72.0,
+      "eval_accuracy": 0.717391304347826,
+      "eval_loss": 0.8566347360610962,
+      "eval_runtime": 0.6595,
+      "eval_samples_per_second": 69.753,
+      "eval_steps_per_second": 3.033,
+      "step": 468
+    },
+    {
+      "epoch": 72.31,
+      "learning_rate": 4.901960784313726e-06,
+      "loss": 0.1033,
+      "step": 470
+    },
+    {
+      "epoch": 72.92,
+      "eval_accuracy": 0.717391304347826,
+      "eval_loss": 0.8909047842025757,
+      "eval_runtime": 0.6783,
+      "eval_samples_per_second": 67.817,
+      "eval_steps_per_second": 2.949,
+      "step": 474
+    },
+    {
+      "epoch": 73.85,
+      "learning_rate": 0.0,
+      "loss": 0.0876,
+      "step": 480
+    },
+    {
+      "epoch": 73.85,
+      "eval_accuracy": 0.7608695652173914,
+      "eval_loss": 0.8869202733039856,
+      "eval_runtime": 0.6277,
+      "eval_samples_per_second": 73.284,
+      "eval_steps_per_second": 3.186,
+      "step": 480
+    },
+    {
+      "epoch": 73.85,
+      "step": 480,
+      "total_flos": 1.6581977329862246e+17,
+      "train_loss": 0.40781121912101903,
+      "train_runtime": 277.6094,
+      "train_samples_per_second": 236.015,
+      "train_steps_per_second": 1.729
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 480,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 80,
+  "save_steps": 500,
+  "total_flos": 1.6581977329862246e+17,
+  "train_batch_size": 32,
+  "trial_name": null,
+  "trial_params": null
+}