End of training

README.md

@@ -22,7 +22,7 @@ model-index:
     metrics:
     - name: Accuracy
       type: accuracy
-      value: 0.8043478260869565
+      value: 0.8695652173913043
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -32,8 +32,8 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [google/vit-base-patch16-224](https://huggingface.co/google/vit-base-patch16-224) on the imagefolder dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.8808
-- Accuracy: 0.8043
+- Loss: 0.6456
+- Accuracy: 0.8696
 
 ## Model description
 

all_results.json

@@ -0,0 +1,12 @@
+{
+    "epoch": 38.83,
+    "eval_accuracy": 0.8695652173913043,
+    "eval_loss": 0.6455863118171692,
+    "eval_runtime": 0.9822,
+    "eval_samples_per_second": 46.833,
+    "eval_steps_per_second": 6.109,
+    "train_loss": 0.2159135382026434,
+    "train_runtime": 637.3873,
+    "train_samples_per_second": 51.397,
+    "train_steps_per_second": 1.569
+}

eval_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 38.83,
+    "eval_accuracy": 0.8695652173913043,
+    "eval_loss": 0.6455863118171692,
+    "eval_runtime": 0.9822,
+    "eval_samples_per_second": 46.833,
+    "eval_steps_per_second": 6.109
+}

runs/Jun23_14-39-12_DESKTOP-SKBE9FB/events.out.tfevents.1719176029.DESKTOP-SKBE9FB.16296.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8e816d8ee1d2541ec4717250e5448c2d816a002efa7103e16a26b553c9dadc15
+size 411

train_results.json

@@ -0,0 +1,7 @@
+{
+    "epoch": 38.83,
+    "train_loss": 0.2159135382026434,
+    "train_runtime": 637.3873,
+    "train_samples_per_second": 51.397,
+    "train_steps_per_second": 1.569
+}

trainer_state.json

@@ -0,0 +1,981 @@
+{
+  "best_metric": 0.8695652173913043,
+  "best_model_checkpoint": "vit-base-patch16-224-ve-U13b-80RX\\checkpoint-540",
+  "epoch": 38.83495145631068,
+  "eval_steps": 500,
+  "global_step": 1000,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.39,
+      "learning_rate": 1.1000000000000001e-05,
+      "loss": 1.386,
+      "step": 10
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 2.2000000000000003e-05,
+      "loss": 1.3805,
+      "step": 20
+    },
+    {
+      "epoch": 0.97,
+      "eval_accuracy": 0.5,
+      "eval_loss": 1.3566619157791138,
+      "eval_runtime": 0.9332,
+      "eval_samples_per_second": 49.292,
+      "eval_steps_per_second": 6.429,
+      "step": 25
+    },
+    {
+      "epoch": 1.17,
+      "learning_rate": 3.3e-05,
+      "loss": 1.3523,
+      "step": 30
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 4.4000000000000006e-05,
+      "loss": 1.283,
+      "step": 40
+    },
+    {
+      "epoch": 1.94,
+      "learning_rate": 5.5e-05,
+      "loss": 1.1576,
+      "step": 50
+    },
+    {
+      "epoch": 1.98,
+      "eval_accuracy": 0.43478260869565216,
+      "eval_loss": 1.1359825134277344,
+      "eval_runtime": 0.7817,
+      "eval_samples_per_second": 58.846,
+      "eval_steps_per_second": 7.676,
+      "step": 51
+    },
+    {
+      "epoch": 2.33,
+      "learning_rate": 5.4421052631578945e-05,
+      "loss": 1.0631,
+      "step": 60
+    },
+    {
+      "epoch": 2.72,
+      "learning_rate": 5.38421052631579e-05,
+      "loss": 0.9331,
+      "step": 70
+    },
+    {
+      "epoch": 2.99,
+      "eval_accuracy": 0.8043478260869565,
+      "eval_loss": 0.8530930876731873,
+      "eval_runtime": 0.8277,
+      "eval_samples_per_second": 55.576,
+      "eval_steps_per_second": 7.249,
+      "step": 77
+    },
+    {
+      "epoch": 3.11,
+      "learning_rate": 5.3263157894736844e-05,
+      "loss": 0.8206,
+      "step": 80
+    },
+    {
+      "epoch": 3.5,
+      "learning_rate": 5.268421052631579e-05,
+      "loss": 0.7139,
+      "step": 90
+    },
+    {
+      "epoch": 3.88,
+      "learning_rate": 5.210526315789474e-05,
+      "loss": 0.6657,
+      "step": 100
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.782608695652174,
+      "eval_loss": 0.6856333613395691,
+      "eval_runtime": 0.7852,
+      "eval_samples_per_second": 58.586,
+      "eval_steps_per_second": 7.642,
+      "step": 103
+    },
+    {
+      "epoch": 4.27,
+      "learning_rate": 5.152631578947369e-05,
+      "loss": 0.6225,
+      "step": 110
+    },
+    {
+      "epoch": 4.66,
+      "learning_rate": 5.0947368421052636e-05,
+      "loss": 0.5642,
+      "step": 120
+    },
+    {
+      "epoch": 4.97,
+      "eval_accuracy": 0.782608695652174,
+      "eval_loss": 0.6162217855453491,
+      "eval_runtime": 0.8007,
+      "eval_samples_per_second": 57.45,
+      "eval_steps_per_second": 7.494,
+      "step": 128
+    },
+    {
+      "epoch": 5.05,
+      "learning_rate": 5.036842105263158e-05,
+      "loss": 0.4835,
+      "step": 130
+    },
+    {
+      "epoch": 5.44,
+      "learning_rate": 4.978947368421053e-05,
+      "loss": 0.4062,
+      "step": 140
+    },
+    {
+      "epoch": 5.83,
+      "learning_rate": 4.921052631578948e-05,
+      "loss": 0.3632,
+      "step": 150
+    },
+    {
+      "epoch": 5.98,
+      "eval_accuracy": 0.8043478260869565,
+      "eval_loss": 0.590190589427948,
+      "eval_runtime": 0.7922,
+      "eval_samples_per_second": 58.068,
+      "eval_steps_per_second": 7.574,
+      "step": 154
+    },
+    {
+      "epoch": 6.21,
+      "learning_rate": 4.863157894736842e-05,
+      "loss": 0.3261,
+      "step": 160
+    },
+    {
+      "epoch": 6.6,
+      "learning_rate": 4.805263157894737e-05,
+      "loss": 0.2892,
+      "step": 170
+    },
+    {
+      "epoch": 6.99,
+      "learning_rate": 4.747368421052632e-05,
+      "loss": 0.3384,
+      "step": 180
+    },
+    {
+      "epoch": 6.99,
+      "eval_accuracy": 0.8043478260869565,
+      "eval_loss": 0.4995403587818146,
+      "eval_runtime": 0.7967,
+      "eval_samples_per_second": 57.739,
+      "eval_steps_per_second": 7.531,
+      "step": 180
+    },
+    {
+      "epoch": 7.38,
+      "learning_rate": 4.6894736842105264e-05,
+      "loss": 0.2762,
+      "step": 190
+    },
+    {
+      "epoch": 7.77,
+      "learning_rate": 4.631578947368421e-05,
+      "loss": 0.2261,
+      "step": 200
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.7608695652173914,
+      "eval_loss": 0.6854467988014221,
+      "eval_runtime": 0.7832,
+      "eval_samples_per_second": 58.734,
+      "eval_steps_per_second": 7.661,
+      "step": 206
+    },
+    {
+      "epoch": 8.16,
+      "learning_rate": 4.573684210526316e-05,
+      "loss": 0.259,
+      "step": 210
+    },
+    {
+      "epoch": 8.54,
+      "learning_rate": 4.5157894736842106e-05,
+      "loss": 0.1924,
+      "step": 220
+    },
+    {
+      "epoch": 8.93,
+      "learning_rate": 4.457894736842105e-05,
+      "loss": 0.2066,
+      "step": 230
+    },
+    {
+      "epoch": 8.97,
+      "eval_accuracy": 0.782608695652174,
+      "eval_loss": 0.5605289936065674,
+      "eval_runtime": 1.1543,
+      "eval_samples_per_second": 39.852,
+      "eval_steps_per_second": 5.198,
+      "step": 231
+    },
+    {
+      "epoch": 9.32,
+      "learning_rate": 4.4000000000000006e-05,
+      "loss": 0.1525,
+      "step": 240
+    },
+    {
+      "epoch": 9.71,
+      "learning_rate": 4.342105263157895e-05,
+      "loss": 0.1635,
+      "step": 250
+    },
+    {
+      "epoch": 9.98,
+      "eval_accuracy": 0.7391304347826086,
+      "eval_loss": 0.720944881439209,
+      "eval_runtime": 0.7752,
+      "eval_samples_per_second": 59.34,
+      "eval_steps_per_second": 7.74,
+      "step": 257
+    },
+    {
+      "epoch": 10.1,
+      "learning_rate": 4.28421052631579e-05,
+      "loss": 0.1457,
+      "step": 260
+    },
+    {
+      "epoch": 10.49,
+      "learning_rate": 4.226315789473684e-05,
+      "loss": 0.1363,
+      "step": 270
+    },
+    {
+      "epoch": 10.87,
+      "learning_rate": 4.168421052631579e-05,
+      "loss": 0.1829,
+      "step": 280
+    },
+    {
+      "epoch": 10.99,
+      "eval_accuracy": 0.6956521739130435,
+      "eval_loss": 0.9292640089988708,
+      "eval_runtime": 0.8392,
+      "eval_samples_per_second": 54.814,
+      "eval_steps_per_second": 7.15,
+      "step": 283
+    },
+    {
+      "epoch": 11.26,
+      "learning_rate": 4.110526315789474e-05,
+      "loss": 0.1439,
+      "step": 290
+    },
+    {
+      "epoch": 11.65,
+      "learning_rate": 4.0526315789473684e-05,
+      "loss": 0.1455,
+      "step": 300
+    },
+    {
+      "epoch": 12.0,
+      "eval_accuracy": 0.782608695652174,
+      "eval_loss": 0.5999487042427063,
+      "eval_runtime": 0.7912,
+      "eval_samples_per_second": 58.141,
+      "eval_steps_per_second": 7.584,
+      "step": 309
+    },
+    {
+      "epoch": 12.04,
+      "learning_rate": 3.994736842105263e-05,
+      "loss": 0.1154,
+      "step": 310
+    },
+    {
+      "epoch": 12.43,
+      "learning_rate": 3.936842105263158e-05,
+      "loss": 0.1461,
+      "step": 320
+    },
+    {
+      "epoch": 12.82,
+      "learning_rate": 3.8789473684210526e-05,
+      "loss": 0.1072,
+      "step": 330
+    },
+    {
+      "epoch": 12.97,
+      "eval_accuracy": 0.782608695652174,
+      "eval_loss": 0.7919036149978638,
+      "eval_runtime": 0.8292,
+      "eval_samples_per_second": 55.476,
+      "eval_steps_per_second": 7.236,
+      "step": 334
+    },
+    {
+      "epoch": 13.2,
+      "learning_rate": 3.8210526315789476e-05,
+      "loss": 0.1154,
+      "step": 340
+    },
+    {
+      "epoch": 13.59,
+      "learning_rate": 3.7631578947368425e-05,
+      "loss": 0.1484,
+      "step": 350
+    },
+    {
+      "epoch": 13.98,
+      "learning_rate": 3.705263157894737e-05,
+      "loss": 0.1059,
+      "step": 360
+    },
+    {
+      "epoch": 13.98,
+      "eval_accuracy": 0.8043478260869565,
+      "eval_loss": 0.7782349586486816,
+      "eval_runtime": 0.7972,
+      "eval_samples_per_second": 57.703,
+      "eval_steps_per_second": 7.526,
+      "step": 360
+    },
+    {
+      "epoch": 14.37,
+      "learning_rate": 3.647368421052631e-05,
+      "loss": 0.0885,
+      "step": 370
+    },
+    {
+      "epoch": 14.76,
+      "learning_rate": 3.589473684210527e-05,
+      "loss": 0.0971,
+      "step": 380
+    },
+    {
+      "epoch": 14.99,
+      "eval_accuracy": 0.8043478260869565,
+      "eval_loss": 0.8248986005783081,
+      "eval_runtime": 0.7922,
+      "eval_samples_per_second": 58.067,
+      "eval_steps_per_second": 7.574,
+      "step": 386
+    },
+    {
+      "epoch": 15.15,
+      "learning_rate": 3.531578947368421e-05,
+      "loss": 0.1269,
+      "step": 390
+    },
+    {
+      "epoch": 15.53,
+      "learning_rate": 3.473684210526316e-05,
+      "loss": 0.1137,
+      "step": 400
+    },
+    {
+      "epoch": 15.92,
+      "learning_rate": 3.415789473684211e-05,
+      "loss": 0.0456,
+      "step": 410
+    },
+    {
+      "epoch": 16.0,
+      "eval_accuracy": 0.782608695652174,
+      "eval_loss": 0.7964589595794678,
+      "eval_runtime": 0.7832,
+      "eval_samples_per_second": 58.735,
+      "eval_steps_per_second": 7.661,
+      "step": 412
+    },
+    {
+      "epoch": 16.31,
+      "learning_rate": 3.357894736842105e-05,
+      "loss": 0.127,
+      "step": 420
+    },
+    {
+      "epoch": 16.7,
+      "learning_rate": 3.3e-05,
+      "loss": 0.0483,
+      "step": 430
+    },
+    {
+      "epoch": 16.97,
+      "eval_accuracy": 0.8260869565217391,
+      "eval_loss": 0.7163321375846863,
+      "eval_runtime": 0.8357,
+      "eval_samples_per_second": 55.044,
+      "eval_steps_per_second": 7.18,
+      "step": 437
+    },
+    {
+      "epoch": 17.09,
+      "learning_rate": 3.2421052631578945e-05,
+      "loss": 0.1415,
+      "step": 440
+    },
+    {
+      "epoch": 17.48,
+      "learning_rate": 3.1842105263157895e-05,
+      "loss": 0.1081,
+      "step": 450
+    },
+    {
+      "epoch": 17.86,
+      "learning_rate": 3.1263157894736845e-05,
+      "loss": 0.0832,
+      "step": 460
+    },
+    {
+      "epoch": 17.98,
+      "eval_accuracy": 0.782608695652174,
+      "eval_loss": 0.8122026324272156,
+      "eval_runtime": 0.7852,
+      "eval_samples_per_second": 58.584,
+      "eval_steps_per_second": 7.641,
+      "step": 463
+    },
+    {
+      "epoch": 18.25,
+      "learning_rate": 3.068421052631579e-05,
+      "loss": 0.0629,
+      "step": 470
+    },
+    {
+      "epoch": 18.64,
+      "learning_rate": 3.010526315789474e-05,
+      "loss": 0.055,
+      "step": 480
+    },
+    {
+      "epoch": 18.99,
+      "eval_accuracy": 0.782608695652174,
+      "eval_loss": 0.8250147104263306,
+      "eval_runtime": 0.7762,
+      "eval_samples_per_second": 59.264,
+      "eval_steps_per_second": 7.73,
+      "step": 489
+    },
+    {
+      "epoch": 19.03,
+      "learning_rate": 2.9526315789473684e-05,
+      "loss": 0.1095,
+      "step": 490
+    },
+    {
+      "epoch": 19.42,
+      "learning_rate": 2.894736842105263e-05,
+      "loss": 0.07,
+      "step": 500
+    },
+    {
+      "epoch": 19.81,
+      "learning_rate": 2.8368421052631583e-05,
+      "loss": 0.0753,
+      "step": 510
+    },
+    {
+      "epoch": 20.0,
+      "eval_accuracy": 0.8478260869565217,
+      "eval_loss": 0.6865554451942444,
+      "eval_runtime": 0.8217,
+      "eval_samples_per_second": 55.981,
+      "eval_steps_per_second": 7.302,
+      "step": 515
+    },
+    {
+      "epoch": 20.19,
+      "learning_rate": 2.778947368421053e-05,
+      "loss": 0.1031,
+      "step": 520
+    },
+    {
+      "epoch": 20.58,
+      "learning_rate": 2.7210526315789472e-05,
+      "loss": 0.0668,
+      "step": 530
+    },
+    {
+      "epoch": 20.97,
+      "learning_rate": 2.6631578947368422e-05,
+      "loss": 0.14,
+      "step": 540
+    },
+    {
+      "epoch": 20.97,
+      "eval_accuracy": 0.8695652173913043,
+      "eval_loss": 0.6455863118171692,
+      "eval_runtime": 0.8212,
+      "eval_samples_per_second": 56.015,
+      "eval_steps_per_second": 7.306,
+      "step": 540
+    },
+    {
+      "epoch": 21.36,
+      "learning_rate": 2.605263157894737e-05,
+      "loss": 0.0812,
+      "step": 550
+    },
+    {
+      "epoch": 21.75,
+      "learning_rate": 2.5473684210526318e-05,
+      "loss": 0.0506,
+      "step": 560
+    },
+    {
+      "epoch": 21.98,
+      "eval_accuracy": 0.782608695652174,
+      "eval_loss": 0.9126870632171631,
+      "eval_runtime": 0.8397,
+      "eval_samples_per_second": 54.781,
+      "eval_steps_per_second": 7.145,
+      "step": 566
+    },
+    {
+      "epoch": 22.14,
+      "learning_rate": 2.4894736842105264e-05,
+      "loss": 0.0413,
+      "step": 570
+    },
+    {
+      "epoch": 22.52,
+      "learning_rate": 2.431578947368421e-05,
+      "loss": 0.1031,
+      "step": 580
+    },
+    {
+      "epoch": 22.91,
+      "learning_rate": 2.373684210526316e-05,
+      "loss": 0.0963,
+      "step": 590
+    },
+    {
+      "epoch": 22.99,
+      "eval_accuracy": 0.8260869565217391,
+      "eval_loss": 0.6365217566490173,
+      "eval_runtime": 0.8112,
+      "eval_samples_per_second": 56.706,
+      "eval_steps_per_second": 7.396,
+      "step": 592
+    },
+    {
+      "epoch": 23.3,
+      "learning_rate": 2.3157894736842103e-05,
+      "loss": 0.0693,
+      "step": 600
+    },
+    {
+      "epoch": 23.69,
+      "learning_rate": 2.2578947368421053e-05,
+      "loss": 0.0612,
+      "step": 610
+    },
+    {
+      "epoch": 24.0,
+      "eval_accuracy": 0.8043478260869565,
+      "eval_loss": 0.8252330422401428,
+      "eval_runtime": 0.8217,
+      "eval_samples_per_second": 55.982,
+      "eval_steps_per_second": 7.302,
+      "step": 618
+    },
+    {
+      "epoch": 24.08,
+      "learning_rate": 2.2000000000000003e-05,
+      "loss": 0.0353,
+      "step": 620
+    },
+    {
+      "epoch": 24.47,
+      "learning_rate": 2.142105263157895e-05,
+      "loss": 0.1415,
+      "step": 630
+    },
+    {
+      "epoch": 24.85,
+      "learning_rate": 2.0842105263157895e-05,
+      "loss": 0.0875,
+      "step": 640
+    },
+    {
+      "epoch": 24.97,
+      "eval_accuracy": 0.7391304347826086,
+      "eval_loss": 0.8843982815742493,
+      "eval_runtime": 0.7792,
+      "eval_samples_per_second": 59.035,
+      "eval_steps_per_second": 7.7,
+      "step": 643
+    },
+    {
+      "epoch": 25.24,
+      "learning_rate": 2.0263157894736842e-05,
+      "loss": 0.1033,
+      "step": 650
+    },
+    {
+      "epoch": 25.63,
+      "learning_rate": 1.968421052631579e-05,
+      "loss": 0.1041,
+      "step": 660
+    },
+    {
+      "epoch": 25.98,
+      "eval_accuracy": 0.8260869565217391,
+      "eval_loss": 0.6594071984291077,
+      "eval_runtime": 0.7782,
+      "eval_samples_per_second": 59.112,
+      "eval_steps_per_second": 7.71,
+      "step": 669
+    },
+    {
+      "epoch": 26.02,
+      "learning_rate": 1.9105263157894738e-05,
+      "loss": 0.0429,
+      "step": 670
+    },
+    {
+      "epoch": 26.41,
+      "learning_rate": 1.8526315789473684e-05,
+      "loss": 0.1199,
+      "step": 680
+    },
+    {
+      "epoch": 26.8,
+      "learning_rate": 1.7947368421052634e-05,
+      "loss": 0.0512,
+      "step": 690
+    },
+    {
+      "epoch": 26.99,
+      "eval_accuracy": 0.782608695652174,
+      "eval_loss": 0.9883273839950562,
+      "eval_runtime": 0.7907,
+      "eval_samples_per_second": 58.177,
+      "eval_steps_per_second": 7.588,
+      "step": 695
+    },
+    {
+      "epoch": 27.18,
+      "learning_rate": 1.736842105263158e-05,
+      "loss": 0.0506,
+      "step": 700
+    },
+    {
+      "epoch": 27.57,
+      "learning_rate": 1.6789473684210526e-05,
+      "loss": 0.0676,
+      "step": 710
+    },
+    {
+      "epoch": 27.96,
+      "learning_rate": 1.6210526315789473e-05,
+      "loss": 0.0675,
+      "step": 720
+    },
+    {
+      "epoch": 28.0,
+      "eval_accuracy": 0.8043478260869565,
+      "eval_loss": 0.9216018915176392,
+      "eval_runtime": 0.8032,
+      "eval_samples_per_second": 57.272,
+      "eval_steps_per_second": 7.47,
+      "step": 721
+    },
+    {
+      "epoch": 28.35,
+      "learning_rate": 1.5631578947368422e-05,
+      "loss": 0.0958,
+      "step": 730
+    },
+    {
+      "epoch": 28.74,
+      "learning_rate": 1.505263157894737e-05,
+      "loss": 0.0492,
+      "step": 740
+    },
+    {
+      "epoch": 28.97,
+      "eval_accuracy": 0.8043478260869565,
+      "eval_loss": 0.9284108877182007,
+      "eval_runtime": 0.7852,
+      "eval_samples_per_second": 58.586,
+      "eval_steps_per_second": 7.642,
+      "step": 746
+    },
+    {
+      "epoch": 29.13,
+      "learning_rate": 1.4473684210526315e-05,
+      "loss": 0.0574,
+      "step": 750
+    },
+    {
+      "epoch": 29.51,
+      "learning_rate": 1.3894736842105265e-05,
+      "loss": 0.0527,
+      "step": 760
+    },
+    {
+      "epoch": 29.9,
+      "learning_rate": 1.3315789473684211e-05,
+      "loss": 0.0679,
+      "step": 770
+    },
+    {
+      "epoch": 29.98,
+      "eval_accuracy": 0.782608695652174,
+      "eval_loss": 0.9341010451316833,
+      "eval_runtime": 0.8282,
+      "eval_samples_per_second": 55.542,
+      "eval_steps_per_second": 7.245,
+      "step": 772
+    },
+    {
+      "epoch": 30.29,
+      "learning_rate": 1.2736842105263159e-05,
+      "loss": 0.0248,
+      "step": 780
+    },
+    {
+      "epoch": 30.68,
+      "learning_rate": 1.2157894736842105e-05,
+      "loss": 0.0996,
+      "step": 790
+    },
+    {
+      "epoch": 30.99,
+      "eval_accuracy": 0.8043478260869565,
+      "eval_loss": 0.9608001112937927,
+      "eval_runtime": 0.8052,
+      "eval_samples_per_second": 57.13,
+      "eval_steps_per_second": 7.452,
+      "step": 798
+    },
+    {
+      "epoch": 31.07,
+      "learning_rate": 1.1578947368421052e-05,
+      "loss": 0.0596,
+      "step": 800
+    },
+    {
+      "epoch": 31.46,
+      "learning_rate": 1.1000000000000001e-05,
+      "loss": 0.047,
+      "step": 810
+    },
+    {
+      "epoch": 31.84,
+      "learning_rate": 1.0421052631578948e-05,
+      "loss": 0.0729,
+      "step": 820
+    },
+    {
+      "epoch": 32.0,
+      "eval_accuracy": 0.782608695652174,
+      "eval_loss": 1.015515685081482,
+      "eval_runtime": 0.7852,
+      "eval_samples_per_second": 58.586,
+      "eval_steps_per_second": 7.642,
+      "step": 824
+    },
+    {
+      "epoch": 32.23,
+      "learning_rate": 9.842105263157896e-06,
+      "loss": 0.0446,
+      "step": 830
+    },
+    {
+      "epoch": 32.62,
+      "learning_rate": 9.263157894736842e-06,
+      "loss": 0.0296,
+      "step": 840
+    },
+    {
+      "epoch": 32.97,
+      "eval_accuracy": 0.782608695652174,
+      "eval_loss": 1.0314323902130127,
+      "eval_runtime": 0.7802,
+      "eval_samples_per_second": 58.96,
+      "eval_steps_per_second": 7.69,
+      "step": 849
+    },
+    {
+      "epoch": 33.01,
+      "learning_rate": 8.68421052631579e-06,
+      "loss": 0.0918,
+      "step": 850
+    },
+    {
+      "epoch": 33.4,
+      "learning_rate": 8.105263157894736e-06,
+      "loss": 0.0642,
+      "step": 860
+    },
+    {
+      "epoch": 33.79,
+      "learning_rate": 7.526315789473685e-06,
+      "loss": 0.0414,
+      "step": 870
+    },
+    {
+      "epoch": 33.98,
+      "eval_accuracy": 0.8043478260869565,
+      "eval_loss": 0.8358449339866638,
+      "eval_runtime": 0.7882,
+      "eval_samples_per_second": 58.362,
+      "eval_steps_per_second": 7.612,
+      "step": 875
+    },
+    {
+      "epoch": 34.17,
+      "learning_rate": 6.947368421052632e-06,
+      "loss": 0.0482,
+      "step": 880
+    },
+    {
+      "epoch": 34.56,
+      "learning_rate": 6.3684210526315795e-06,
+      "loss": 0.0462,
+      "step": 890
+    },
+    {
+      "epoch": 34.95,
+      "learning_rate": 5.789473684210526e-06,
+      "loss": 0.04,
+      "step": 900
+    },
+    {
+      "epoch": 34.99,
+      "eval_accuracy": 0.8043478260869565,
+      "eval_loss": 0.891183614730835,
+      "eval_runtime": 0.8252,
+      "eval_samples_per_second": 55.745,
+      "eval_steps_per_second": 7.271,
+      "step": 901
+    },
+    {
+      "epoch": 35.34,
+      "learning_rate": 5.210526315789474e-06,
+      "loss": 0.0262,
+      "step": 910
+    },
+    {
+      "epoch": 35.73,
+      "learning_rate": 4.631578947368421e-06,
+      "loss": 0.0179,
+      "step": 920
+    },
+    {
+      "epoch": 36.0,
+      "eval_accuracy": 0.8043478260869565,
+      "eval_loss": 0.8544241786003113,
+      "eval_runtime": 0.7722,
+      "eval_samples_per_second": 59.571,
+      "eval_steps_per_second": 7.77,
+      "step": 927
+    },
+    {
+      "epoch": 36.12,
+      "learning_rate": 4.052631578947368e-06,
+      "loss": 0.0205,
+      "step": 930
+    },
+    {
+      "epoch": 36.5,
+      "learning_rate": 3.473684210526316e-06,
+      "loss": 0.0586,
+      "step": 940
+    },
+    {
+      "epoch": 36.89,
+      "learning_rate": 2.894736842105263e-06,
+      "loss": 0.0665,
+      "step": 950
+    },
+    {
+      "epoch": 36.97,
+      "eval_accuracy": 0.8043478260869565,
+      "eval_loss": 0.9153819680213928,
+      "eval_runtime": 0.7792,
+      "eval_samples_per_second": 59.037,
+      "eval_steps_per_second": 7.7,
+      "step": 952
+    },
+    {
+      "epoch": 37.28,
+      "learning_rate": 2.3157894736842105e-06,
+      "loss": 0.0597,
+      "step": 960
+    },
+    {
+      "epoch": 37.67,
+      "learning_rate": 1.736842105263158e-06,
+      "loss": 0.0413,
+      "step": 970
+    },
+    {
+      "epoch": 37.98,
+      "eval_accuracy": 0.8043478260869565,
+      "eval_loss": 0.8834414482116699,
+      "eval_runtime": 0.8242,
+      "eval_samples_per_second": 55.81,
+      "eval_steps_per_second": 7.28,
+      "step": 978
+    },
+    {
+      "epoch": 38.06,
+      "learning_rate": 1.1578947368421053e-06,
+      "loss": 0.0218,
+      "step": 980
+    },
+    {
+      "epoch": 38.45,
+      "learning_rate": 5.789473684210526e-07,
+      "loss": 0.0512,
+      "step": 990
+    },
+    {
+      "epoch": 38.83,
+      "learning_rate": 0.0,
+      "loss": 0.04,
+      "step": 1000
+    },
+    {
+      "epoch": 38.83,
+      "eval_accuracy": 0.8043478260869565,
+      "eval_loss": 0.8807706236839294,
+      "eval_runtime": 0.8612,
+      "eval_samples_per_second": 53.414,
+      "eval_steps_per_second": 6.967,
+      "step": 1000
+    },
+    {
+      "epoch": 38.83,
+      "step": 1000,
+      "total_flos": 2.4650643760327066e+18,
+      "train_loss": 0.2159135382026434,
+      "train_runtime": 637.3873,
+      "train_samples_per_second": 51.397,
+      "train_steps_per_second": 1.569
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 1000,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 40,
+  "save_steps": 500,
+  "total_flos": 2.4650643760327066e+18,
+  "train_batch_size": 8,
+  "trial_name": null,
+  "trial_params": null
+}