End of training

README.md

@@ -22,7 +22,7 @@ model-index:
     metrics:
     - name: Accuracy
       type: accuracy
-      value: 0.8985148514851485
+      value: 0.9034653465346535
 ---
 
 <!-- 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 [Safawat/Electrcical-IMAGE-finetuned-eurosat](https://huggingface.co/Safawat/Electrcical-IMAGE-finetuned-eurosat) on the imagefolder dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.3104
-- Accuracy: 0.8985
+- Loss: 0.3388
+- Accuracy: 0.9035
 
 ## Model description
 

all_results.json

@@ -1,13 +1,13 @@
 {
-    "epoch": 9.824561403508772,
-    "eval_accuracy": 0.8910891089108911,
-    "eval_loss": 0.3172283470630646,
-    "eval_runtime": 4.4319,
-    "eval_samples_per_second": 91.157,
-    "eval_steps_per_second": 2.933,
-    "total_flos": 8.878060093031055e+17,
-    "train_loss": 0.5355421313217708,
-    "train_runtime": 1702.5623,
-    "train_samples_per_second": 21.344,
-    "train_steps_per_second": 0.164
+    "epoch": 19.649122807017545,
+    "eval_accuracy": 0.9034653465346535,
+    "eval_loss": 0.33881905674934387,
+    "eval_runtime": 4.2245,
+    "eval_samples_per_second": 95.632,
+    "eval_steps_per_second": 3.077,
+    "total_flos": 1.7752639958663823e+18,
+    "train_loss": 0.17112446086747304,
+    "train_runtime": 1440.5794,
+    "train_samples_per_second": 50.452,
+    "train_steps_per_second": 0.389
 }

eval_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 9.824561403508772,
-    "eval_accuracy": 0.8910891089108911,
-    "eval_loss": 0.3172283470630646,
-    "eval_runtime": 4.4319,
-    "eval_samples_per_second": 91.157,
-    "eval_steps_per_second": 2.933
+    "epoch": 19.649122807017545,
+    "eval_accuracy": 0.9034653465346535,
+    "eval_loss": 0.33881905674934387,
+    "eval_runtime": 4.2245,
+    "eval_samples_per_second": 95.632,
+    "eval_steps_per_second": 3.077
 }

runs/Apr26_02-53-40_0fb91cab2fc3/events.out.tfevents.1714101478.0fb91cab2fc3.5439.4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9837cdc38adc7c654fbe4b6f572d573ed37d1d30cc48bf3523cc10c0d9c5c1fb
+size 411

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 9.824561403508772,
-    "total_flos": 8.878060093031055e+17,
-    "train_loss": 0.5355421313217708,
-    "train_runtime": 1702.5623,
-    "train_samples_per_second": 21.344,
-    "train_steps_per_second": 0.164
+    "epoch": 19.649122807017545,
+    "total_flos": 1.7752639958663823e+18,
+    "train_loss": 0.17112446086747304,
+    "train_runtime": 1440.5794,
+    "train_samples_per_second": 50.452,
+    "train_steps_per_second": 0.389
 }

trainer_state.json

@@ -1,315 +1,601 @@
 {
-  "best_metric": 0.8910891089108911,
-  "best_model_checkpoint": "Electrcical-IMAGE-finetuned-eurosat/checkpoint-199",
-  "epoch": 9.824561403508772,
+  "best_metric": 0.9034653465346535,
+  "best_model_checkpoint": "Electrcical-IMAGE-finetuned-eurosat/checkpoint-171",
+  "epoch": 19.649122807017545,
   "eval_steps": 500,
-  "global_step": 280,
+  "global_step": 560,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
       "epoch": 0.3508771929824561,
-      "grad_norm": 10.464795112609863,
-      "learning_rate": 1.785714285714286e-05,
-      "loss": 1.7823,
+      "grad_norm": 6.915553569793701,
+      "learning_rate": 8.92857142857143e-06,
+      "loss": 0.2535,
       "step": 10
     },
     {
       "epoch": 0.7017543859649122,
-      "grad_norm": 9.879911422729492,
-      "learning_rate": 3.571428571428572e-05,
-      "loss": 1.4879,
+      "grad_norm": 6.068195819854736,
+      "learning_rate": 1.785714285714286e-05,
+      "loss": 0.2536,
       "step": 20
     },
     {
       "epoch": 0.9824561403508771,
-      "eval_accuracy": 0.7326732673267327,
-      "eval_loss": 0.9158226847648621,
-      "eval_runtime": 116.3827,
-      "eval_samples_per_second": 3.471,
-      "eval_steps_per_second": 0.112,
+      "eval_accuracy": 0.8985148514851485,
+      "eval_loss": 0.3073369264602661,
+      "eval_runtime": 4.5045,
+      "eval_samples_per_second": 89.687,
+      "eval_steps_per_second": 2.886,
       "step": 28
     },
     {
       "epoch": 1.0526315789473684,
-      "grad_norm": 7.517858028411865,
-      "learning_rate": 4.960317460317461e-05,
-      "loss": 1.1987,
+      "grad_norm": 4.642627716064453,
+      "learning_rate": 2.6785714285714288e-05,
+      "loss": 0.2506,
       "step": 30
     },
     {
       "epoch": 1.4035087719298245,
-      "grad_norm": 7.376903533935547,
-      "learning_rate": 4.761904761904762e-05,
-      "loss": 0.8925,
+      "grad_norm": 5.193023681640625,
+      "learning_rate": 3.571428571428572e-05,
+      "loss": 0.2254,
       "step": 40
     },
     {
       "epoch": 1.7543859649122808,
-      "grad_norm": 8.916149139404297,
-      "learning_rate": 4.563492063492064e-05,
-      "loss": 0.7072,
+      "grad_norm": 6.378445148468018,
+      "learning_rate": 4.464285714285715e-05,
+      "loss": 0.1702,
       "step": 50
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.8366336633663366,
-      "eval_loss": 0.4648021161556244,
-      "eval_runtime": 4.3367,
-      "eval_samples_per_second": 93.158,
-      "eval_steps_per_second": 2.998,
+      "eval_accuracy": 0.8811881188118812,
+      "eval_loss": 0.32594048976898193,
+      "eval_runtime": 4.1579,
+      "eval_samples_per_second": 97.165,
+      "eval_steps_per_second": 3.127,
       "step": 57
     },
     {
       "epoch": 2.1052631578947367,
-      "grad_norm": 8.09045124053955,
-      "learning_rate": 4.3650793650793655e-05,
-      "loss": 0.665,
+      "grad_norm": 6.762551307678223,
+      "learning_rate": 4.960317460317461e-05,
+      "loss": 0.2028,
       "step": 60
     },
     {
       "epoch": 2.456140350877193,
-      "grad_norm": 9.807299613952637,
-      "learning_rate": 4.166666666666667e-05,
-      "loss": 0.5713,
+      "grad_norm": 9.539087295532227,
+      "learning_rate": 4.8611111111111115e-05,
+      "loss": 0.166,
       "step": 70
     },
     {
       "epoch": 2.807017543859649,
-      "grad_norm": 10.828336715698242,
-      "learning_rate": 3.968253968253968e-05,
-      "loss": 0.521,
+      "grad_norm": 5.8399810791015625,
+      "learning_rate": 4.761904761904762e-05,
+      "loss": 0.1528,
       "step": 80
     },
     {
       "epoch": 2.982456140350877,
-      "eval_accuracy": 0.8712871287128713,
-      "eval_loss": 0.3816491961479187,
-      "eval_runtime": 4.1894,
-      "eval_samples_per_second": 96.434,
-      "eval_steps_per_second": 3.103,
+      "eval_accuracy": 0.8836633663366337,
+      "eval_loss": 0.3369888365268707,
+      "eval_runtime": 4.4971,
+      "eval_samples_per_second": 89.837,
+      "eval_steps_per_second": 2.891,
       "step": 85
     },
     {
       "epoch": 3.1578947368421053,
-      "grad_norm": 6.668788433074951,
-      "learning_rate": 3.76984126984127e-05,
-      "loss": 0.5152,
+      "grad_norm": 2.9166860580444336,
+      "learning_rate": 4.662698412698413e-05,
+      "loss": 0.1767,
       "step": 90
     },
     {
       "epoch": 3.5087719298245617,
-      "grad_norm": 5.836483478546143,
-      "learning_rate": 3.571428571428572e-05,
-      "loss": 0.473,
+      "grad_norm": 3.477200984954834,
+      "learning_rate": 4.563492063492064e-05,
+      "loss": 0.1722,
       "step": 100
     },
     {
       "epoch": 3.8596491228070176,
-      "grad_norm": 5.453497409820557,
-      "learning_rate": 3.3730158730158734e-05,
-      "loss": 0.4664,
+      "grad_norm": 3.7534408569335938,
+      "learning_rate": 4.464285714285715e-05,
+      "loss": 0.157,
       "step": 110
     },
     {
       "epoch": 4.0,
-      "eval_accuracy": 0.8564356435643564,
-      "eval_loss": 0.40334010124206543,
-      "eval_runtime": 4.398,
-      "eval_samples_per_second": 91.86,
-      "eval_steps_per_second": 2.956,
+      "eval_accuracy": 0.8811881188118812,
+      "eval_loss": 0.3372877538204193,
+      "eval_runtime": 4.1256,
+      "eval_samples_per_second": 97.926,
+      "eval_steps_per_second": 3.151,
       "step": 114
     },
     {
       "epoch": 4.2105263157894735,
-      "grad_norm": 6.095137596130371,
-      "learning_rate": 3.1746031746031745e-05,
-      "loss": 0.3831,
+      "grad_norm": 5.6130452156066895,
+      "learning_rate": 4.3650793650793655e-05,
+      "loss": 0.1275,
       "step": 120
     },
     {
       "epoch": 4.56140350877193,
-      "grad_norm": 6.058220386505127,
-      "learning_rate": 2.9761904761904762e-05,
-      "loss": 0.4376,
+      "grad_norm": 6.323825836181641,
+      "learning_rate": 4.265873015873016e-05,
+      "loss": 0.1678,
       "step": 130
     },
     {
       "epoch": 4.912280701754386,
-      "grad_norm": 6.937771797180176,
-      "learning_rate": 2.777777777777778e-05,
-      "loss": 0.3944,
+      "grad_norm": 8.424270629882812,
+      "learning_rate": 4.166666666666667e-05,
+      "loss": 0.1577,
       "step": 140
     },
     {
       "epoch": 4.982456140350877,
-      "eval_accuracy": 0.8737623762376238,
-      "eval_loss": 0.3690718114376068,
-      "eval_runtime": 4.2858,
-      "eval_samples_per_second": 94.264,
-      "eval_steps_per_second": 3.033,
+      "eval_accuracy": 0.8861386138613861,
+      "eval_loss": 0.3574221134185791,
+      "eval_runtime": 4.1778,
+      "eval_samples_per_second": 96.701,
+      "eval_steps_per_second": 3.112,
       "step": 142
     },
     {
       "epoch": 5.2631578947368425,
-      "grad_norm": 6.101373672485352,
-      "learning_rate": 2.5793650793650796e-05,
-      "loss": 0.3582,
+      "grad_norm": 7.5651936531066895,
+      "learning_rate": 4.067460317460318e-05,
+      "loss": 0.1328,
       "step": 150
     },
     {
       "epoch": 5.614035087719298,
-      "grad_norm": 8.85653305053711,
-      "learning_rate": 2.380952380952381e-05,
-      "loss": 0.3584,
+      "grad_norm": 7.6120195388793945,
+      "learning_rate": 3.968253968253968e-05,
+      "loss": 0.1428,
       "step": 160
     },
     {
       "epoch": 5.964912280701754,
-      "grad_norm": 7.283915996551514,
-      "learning_rate": 2.1825396825396827e-05,
-      "loss": 0.3627,
+      "grad_norm": 9.945347785949707,
+      "learning_rate": 3.8690476190476195e-05,
+      "loss": 0.1485,
       "step": 170
     },
     {
       "epoch": 6.0,
-      "eval_accuracy": 0.8886138613861386,
-      "eval_loss": 0.3214375674724579,
-      "eval_runtime": 4.5245,
-      "eval_samples_per_second": 89.292,
-      "eval_steps_per_second": 2.873,
+      "eval_accuracy": 0.9034653465346535,
+      "eval_loss": 0.33881905674934387,
+      "eval_runtime": 4.2693,
+      "eval_samples_per_second": 94.629,
+      "eval_steps_per_second": 3.045,
       "step": 171
     },
     {
       "epoch": 6.315789473684211,
-      "grad_norm": 5.079178810119629,
-      "learning_rate": 1.984126984126984e-05,
-      "loss": 0.3304,
+      "grad_norm": 5.8488969802856445,
+      "learning_rate": 3.76984126984127e-05,
+      "loss": 0.1419,
       "step": 180
     },
     {
       "epoch": 6.666666666666667,
-      "grad_norm": 5.342247486114502,
-      "learning_rate": 1.785714285714286e-05,
-      "loss": 0.3298,
+      "grad_norm": 5.49446964263916,
+      "learning_rate": 3.6706349206349205e-05,
+      "loss": 0.1715,
       "step": 190
     },
     {
       "epoch": 6.982456140350877,
-      "eval_accuracy": 0.8910891089108911,
-      "eval_loss": 0.3172283470630646,
-      "eval_runtime": 4.1397,
-      "eval_samples_per_second": 97.593,
-      "eval_steps_per_second": 3.14,
+      "eval_accuracy": 0.8935643564356436,
+      "eval_loss": 0.3339557945728302,
+      "eval_runtime": 4.2324,
+      "eval_samples_per_second": 95.453,
+      "eval_steps_per_second": 3.072,
       "step": 199
     },
     {
       "epoch": 7.017543859649122,
-      "grad_norm": 6.236889362335205,
-      "learning_rate": 1.5873015873015872e-05,
-      "loss": 0.342,
+      "grad_norm": 7.122160911560059,
+      "learning_rate": 3.571428571428572e-05,
+      "loss": 0.2154,
       "step": 200
     },
     {
       "epoch": 7.368421052631579,
-      "grad_norm": 9.212471008300781,
-      "learning_rate": 1.388888888888889e-05,
-      "loss": 0.3288,
+      "grad_norm": 7.540256977081299,
+      "learning_rate": 3.472222222222222e-05,
+      "loss": 0.2314,
       "step": 210
     },
     {
       "epoch": 7.719298245614035,
-      "grad_norm": 5.810153484344482,
-      "learning_rate": 1.1904761904761905e-05,
-      "loss": 0.3203,
+      "grad_norm": 6.774203777313232,
+      "learning_rate": 3.3730158730158734e-05,
+      "loss": 0.2079,
       "step": 220
     },
     {
       "epoch": 8.0,
-      "eval_accuracy": 0.8910891089108911,
-      "eval_loss": 0.3060537278652191,
-      "eval_runtime": 4.5988,
-      "eval_samples_per_second": 87.848,
-      "eval_steps_per_second": 2.827,
+      "eval_accuracy": 0.8960396039603961,
+      "eval_loss": 0.3254563808441162,
+      "eval_runtime": 4.306,
+      "eval_samples_per_second": 93.823,
+      "eval_steps_per_second": 3.019,
       "step": 228
     },
     {
       "epoch": 8.070175438596491,
-      "grad_norm": 5.650562763214111,
-      "learning_rate": 9.92063492063492e-06,
-      "loss": 0.3379,
+      "grad_norm": 8.224860191345215,
+      "learning_rate": 3.273809523809524e-05,
+      "loss": 0.223,
       "step": 230
     },
     {
       "epoch": 8.421052631578947,
-      "grad_norm": 3.9896435737609863,
-      "learning_rate": 7.936507936507936e-06,
-      "loss": 0.2913,
+      "grad_norm": 4.823063850402832,
+      "learning_rate": 3.1746031746031745e-05,
+      "loss": 0.1889,
       "step": 240
     },
     {
       "epoch": 8.771929824561404,
-      "grad_norm": 7.3288397789001465,
-      "learning_rate": 5.9523809523809525e-06,
-      "loss": 0.2737,
+      "grad_norm": 6.559729099273682,
+      "learning_rate": 3.075396825396826e-05,
+      "loss": 0.1846,
       "step": 250
     },
     {
       "epoch": 8.982456140350877,
-      "eval_accuracy": 0.8861386138613861,
-      "eval_loss": 0.3128886818885803,
-      "eval_runtime": 4.2212,
-      "eval_samples_per_second": 95.706,
-      "eval_steps_per_second": 3.08,
+      "eval_accuracy": 0.8985148514851485,
+      "eval_loss": 0.33677831292152405,
+      "eval_runtime": 4.1555,
+      "eval_samples_per_second": 97.221,
+      "eval_steps_per_second": 3.128,
       "step": 256
     },
     {
       "epoch": 9.12280701754386,
-      "grad_norm": 5.5674285888671875,
-      "learning_rate": 3.968253968253968e-06,
-      "loss": 0.2991,
+      "grad_norm": 4.693236827850342,
+      "learning_rate": 2.9761904761904762e-05,
+      "loss": 0.1898,
       "step": 260
     },
     {
       "epoch": 9.473684210526315,
-      "grad_norm": 6.089910984039307,
-      "learning_rate": 1.984126984126984e-06,
-      "loss": 0.2943,
+      "grad_norm": 7.4912261962890625,
+      "learning_rate": 2.876984126984127e-05,
+      "loss": 0.2097,
       "step": 270
     },
     {
       "epoch": 9.824561403508772,
-      "grad_norm": 5.232763290405273,
-      "learning_rate": 0.0,
-      "loss": 0.2728,
+      "grad_norm": 5.2636213302612305,
+      "learning_rate": 2.777777777777778e-05,
+      "loss": 0.1829,
       "step": 280
     },
     {
-      "epoch": 9.824561403508772,
-      "eval_accuracy": 0.8861386138613861,
-      "eval_loss": 0.308758020401001,
-      "eval_runtime": 4.3884,
-      "eval_samples_per_second": 92.061,
-      "eval_steps_per_second": 2.962,
-      "step": 280
+      "epoch": 10.0,
+      "eval_accuracy": 0.8985148514851485,
+      "eval_loss": 0.317474901676178,
+      "eval_runtime": 4.3916,
+      "eval_samples_per_second": 91.994,
+      "eval_steps_per_second": 2.96,
+      "step": 285
     },
     {
-      "epoch": 9.824561403508772,
-      "step": 280,
-      "total_flos": 8.878060093031055e+17,
-      "train_loss": 0.5355421313217708,
-      "train_runtime": 1702.5623,
-      "train_samples_per_second": 21.344,
-      "train_steps_per_second": 0.164
+      "epoch": 10.175438596491228,
+      "grad_norm": 4.9916558265686035,
+      "learning_rate": 2.6785714285714288e-05,
+      "loss": 0.1684,
+      "step": 290
+    },
+    {
+      "epoch": 10.526315789473685,
+      "grad_norm": 6.921645641326904,
+      "learning_rate": 2.5793650793650796e-05,
+      "loss": 0.1897,
+      "step": 300
+    },
+    {
+      "epoch": 10.87719298245614,
+      "grad_norm": 9.2024507522583,
+      "learning_rate": 2.4801587301587305e-05,
+      "loss": 0.1874,
+      "step": 310
+    },
+    {
+      "epoch": 10.982456140350877,
+      "eval_accuracy": 0.8910891089108911,
+      "eval_loss": 0.3384435176849365,
+      "eval_runtime": 4.1888,
+      "eval_samples_per_second": 96.449,
+      "eval_steps_per_second": 3.104,
+      "step": 313
+    },
+    {
+      "epoch": 11.228070175438596,
+      "grad_norm": 3.4831368923187256,
+      "learning_rate": 2.380952380952381e-05,
+      "loss": 0.1468,
+      "step": 320
+    },
+    {
+      "epoch": 11.578947368421053,
+      "grad_norm": 30.23830795288086,
+      "learning_rate": 2.281746031746032e-05,
+      "loss": 0.1723,
+      "step": 330
+    },
+    {
+      "epoch": 11.929824561403509,
+      "grad_norm": 6.060935974121094,
+      "learning_rate": 2.1825396825396827e-05,
+      "loss": 0.1826,
+      "step": 340
+    },
+    {
+      "epoch": 12.0,
+      "eval_accuracy": 0.8985148514851485,
+      "eval_loss": 0.32836616039276123,
+      "eval_runtime": 4.5565,
+      "eval_samples_per_second": 88.664,
+      "eval_steps_per_second": 2.853,
+      "step": 342
+    },
+    {
+      "epoch": 12.280701754385966,
+      "grad_norm": 8.074810981750488,
+      "learning_rate": 2.0833333333333336e-05,
+      "loss": 0.1798,
+      "step": 350
+    },
+    {
+      "epoch": 12.631578947368421,
+      "grad_norm": 6.300619125366211,
+      "learning_rate": 1.984126984126984e-05,
+      "loss": 0.1516,
+      "step": 360
+    },
+    {
+      "epoch": 12.982456140350877,
+      "grad_norm": 8.272747993469238,
+      "learning_rate": 1.884920634920635e-05,
+      "loss": 0.1911,
+      "step": 370
+    },
+    {
+      "epoch": 12.982456140350877,
+      "eval_accuracy": 0.8910891089108911,
+      "eval_loss": 0.3503906726837158,
+      "eval_runtime": 4.1749,
+      "eval_samples_per_second": 96.769,
+      "eval_steps_per_second": 3.114,
+      "step": 370
+    },
+    {
+      "epoch": 13.333333333333334,
+      "grad_norm": 6.439492702484131,
+      "learning_rate": 1.785714285714286e-05,
+      "loss": 0.1755,
+      "step": 380
+    },
+    {
+      "epoch": 13.68421052631579,
+      "grad_norm": 10.013567924499512,
+      "learning_rate": 1.6865079365079367e-05,
+      "loss": 0.1552,
+      "step": 390
+    },
+    {
+      "epoch": 14.0,
+      "eval_accuracy": 0.8935643564356436,
+      "eval_loss": 0.31156373023986816,
+      "eval_runtime": 4.4403,
+      "eval_samples_per_second": 90.985,
+      "eval_steps_per_second": 2.928,
+      "step": 399
+    },
+    {
+      "epoch": 14.035087719298245,
+      "grad_norm": 4.990965843200684,
+      "learning_rate": 1.5873015873015872e-05,
+      "loss": 0.1737,
+      "step": 400
+    },
+    {
+      "epoch": 14.385964912280702,
+      "grad_norm": 5.88447380065918,
+      "learning_rate": 1.4880952380952381e-05,
+      "loss": 0.1589,
+      "step": 410
+    },
+    {
+      "epoch": 14.736842105263158,
+      "grad_norm": 7.048079490661621,
+      "learning_rate": 1.388888888888889e-05,
+      "loss": 0.1855,
+      "step": 420
+    },
+    {
+      "epoch": 14.982456140350877,
+      "eval_accuracy": 0.8960396039603961,
+      "eval_loss": 0.3131696581840515,
+      "eval_runtime": 4.1515,
+      "eval_samples_per_second": 97.315,
+      "eval_steps_per_second": 3.131,
+      "step": 427
+    },
+    {
+      "epoch": 15.087719298245615,
+      "grad_norm": 5.223410129547119,
+      "learning_rate": 1.2896825396825398e-05,
+      "loss": 0.1491,
+      "step": 430
+    },
+    {
+      "epoch": 15.43859649122807,
+      "grad_norm": 5.402275562286377,
+      "learning_rate": 1.1904761904761905e-05,
+      "loss": 0.1627,
+      "step": 440
+    },
+    {
+      "epoch": 15.789473684210526,
+      "grad_norm": 8.536218643188477,
+      "learning_rate": 1.0912698412698414e-05,
+      "loss": 0.1537,
+      "step": 450
+    },
+    {
+      "epoch": 16.0,
+      "eval_accuracy": 0.900990099009901,
+      "eval_loss": 0.30002227425575256,
+      "eval_runtime": 4.5205,
+      "eval_samples_per_second": 89.37,
+      "eval_steps_per_second": 2.876,
+      "step": 456
+    },
+    {
+      "epoch": 16.140350877192983,
+      "grad_norm": 6.947802543640137,
+      "learning_rate": 9.92063492063492e-06,
+      "loss": 0.1511,
+      "step": 460
+    },
+    {
+      "epoch": 16.49122807017544,
+      "grad_norm": 2.9147555828094482,
+      "learning_rate": 8.92857142857143e-06,
+      "loss": 0.1232,
+      "step": 470
+    },
+    {
+      "epoch": 16.842105263157894,
+      "grad_norm": 4.058919906616211,
+      "learning_rate": 7.936507936507936e-06,
+      "loss": 0.118,
+      "step": 480
+    },
+    {
+      "epoch": 16.982456140350877,
+      "eval_accuracy": 0.8985148514851485,
+      "eval_loss": 0.326092928647995,
+      "eval_runtime": 4.138,
+      "eval_samples_per_second": 97.632,
+      "eval_steps_per_second": 3.142,
+      "step": 484
+    },
+    {
+      "epoch": 17.19298245614035,
+      "grad_norm": 8.858957290649414,
+      "learning_rate": 6.944444444444445e-06,
+      "loss": 0.1479,
+      "step": 490
+    },
+    {
+      "epoch": 17.54385964912281,
+      "grad_norm": 6.7740373611450195,
+      "learning_rate": 5.9523809523809525e-06,
+      "loss": 0.1512,
+      "step": 500
+    },
+    {
+      "epoch": 17.894736842105264,
+      "grad_norm": 5.1415534019470215,
+      "learning_rate": 4.96031746031746e-06,
+      "loss": 0.1483,
+      "step": 510
+    },
+    {
+      "epoch": 18.0,
+      "eval_accuracy": 0.8960396039603961,
+      "eval_loss": 0.31427323818206787,
+      "eval_runtime": 4.8063,
+      "eval_samples_per_second": 84.056,
+      "eval_steps_per_second": 2.705,
+      "step": 513
+    },
+    {
+      "epoch": 18.24561403508772,
+      "grad_norm": 5.084235668182373,
+      "learning_rate": 3.968253968253968e-06,
+      "loss": 0.1392,
+      "step": 520
+    },
+    {
+      "epoch": 18.596491228070175,
+      "grad_norm": 4.290576934814453,
+      "learning_rate": 2.9761904761904763e-06,
+      "loss": 0.1191,
+      "step": 530
+    },
+    {
+      "epoch": 18.94736842105263,
+      "grad_norm": 5.261678218841553,
+      "learning_rate": 1.984126984126984e-06,
+      "loss": 0.1169,
+      "step": 540
+    },
+    {
+      "epoch": 18.982456140350877,
+      "eval_accuracy": 0.9034653465346535,
+      "eval_loss": 0.3101979196071625,
+      "eval_runtime": 4.2151,
+      "eval_samples_per_second": 95.846,
+      "eval_steps_per_second": 3.084,
+      "step": 541
+    },
+    {
+      "epoch": 19.29824561403509,
+      "grad_norm": 5.632396221160889,
+      "learning_rate": 9.92063492063492e-07,
+      "loss": 0.1521,
+      "step": 550
+    },
+    {
+      "epoch": 19.649122807017545,
+      "grad_norm": 6.291007995605469,
+      "learning_rate": 0.0,
+      "loss": 0.1268,
+      "step": 560
+    },
+    {
+      "epoch": 19.649122807017545,
+      "eval_accuracy": 0.8985148514851485,
+      "eval_loss": 0.3104223906993866,
+      "eval_runtime": 4.1976,
+      "eval_samples_per_second": 96.244,
+      "eval_steps_per_second": 3.097,
+      "step": 560
+    },
+    {
+      "epoch": 19.649122807017545,
+      "step": 560,
+      "total_flos": 1.7752639958663823e+18,
+      "train_loss": 0.17112446086747304,
+      "train_runtime": 1440.5794,
+      "train_samples_per_second": 50.452,
+      "train_steps_per_second": 0.389
     }
   ],
   "logging_steps": 10,
-  "max_steps": 280,
+  "max_steps": 560,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 10,
+  "num_train_epochs": 20,
   "save_steps": 500,
-  "total_flos": 8.878060093031055e+17,
+  "total_flos": 1.7752639958663823e+18,
   "train_batch_size": 32,
   "trial_name": null,
   "trial_params": null