Model save

README.md

@@ -3,29 +3,12 @@ library_name: transformers
 license: apache-2.0
 base_model: google/vit-base-patch16-224-in21k
 tags:
-- image-classification
-- vision
 - generated_from_trainer
-datasets:
-- imagefolder
 metrics:
 - accuracy
 model-index:
 - name: vit-base-beans
-  results:
-  - task:
-      name: Image Classification
-      type: image-classification
-    dataset:
-      name: imagefolder
-      type: imagefolder
-      config: default
-      split: validation
-      args: default
-    metrics:
-    - name: Accuracy
-      type: accuracy
-      value: 0.9924812030075187
+  results: []
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -33,10 +16,10 @@ should probably proofread and complete it, then remove this comment. -->
 
 # vit-base-beans
 
-This model is a fine-tuned version of [google/vit-base-patch16-224-in21k](https://huggingface.co/google/vit-base-patch16-224-in21k) on the imagefolder dataset.
+This model is a fine-tuned version of [google/vit-base-patch16-224-in21k](https://huggingface.co/google/vit-base-patch16-224-in21k) on an unknown dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.0650
-- Accuracy: 0.9925
+- Accuracy: 0.9774
+- Loss: 0.0817
 
 ## Model description
 
@@ -59,6 +42,10 @@ The following hyperparameters were used during training:
 - train_batch_size: 8
 - eval_batch_size: 8
 - seed: 1337
+- distributed_type: multi-GPU
+- num_devices: 2
+- total_train_batch_size: 16
+- total_eval_batch_size: 16
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
 - num_epochs: 5.0
@@ -67,11 +54,11 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch | Step | Accuracy | Validation Loss |
 |:-------------:|:-----:|:----:|:--------:|:---------------:|
-| 0.2815        | 1.0   | 130  | 0.9624   | 0.2165          |
-| 0.1287        | 2.0   | 260  | 0.9699   | 0.1313          |
-| 0.1408        | 3.0   | 390  | 0.9699   | 0.0998          |
-| 0.0846        | 4.0   | 520  | 0.9925   | 0.0650          |
-| 0.1105        | 5.0   | 650  | 0.9774   | 0.0837          |
+| 0.2859        | 1.0   | 130  | 0.9624   | 0.2189          |
+| 0.1316        | 2.0   | 260  | 0.9699   | 0.1334          |
+| 0.1438        | 3.0   | 390  | 0.9699   | 0.0981          |
+| 0.0833        | 4.0   | 520  | 0.9925   | 0.0656          |
+| 0.1107        | 5.0   | 650  | 0.9774   | 0.0817          |
 
 
 ### Framework versions

all_results.json

@@ -1,13 +1,13 @@
 {
     "epoch": 5.0,
     "eval_accuracy": 0.9924812030075187,
-    "eval_loss": 0.06499230116605759,
-    "eval_runtime": 1.0883,
-    "eval_samples_per_second": 122.212,
-    "eval_steps_per_second": 15.621,
+    "eval_loss": 0.06558172404766083,
+    "eval_runtime": 1.0482,
+    "eval_samples_per_second": 126.888,
+    "eval_steps_per_second": 16.219,
     "total_flos": 4.006371770595533e+17,
-    "train_loss": 0.0,
-    "train_runtime": 0.0792,
-    "train_samples_per_second": 65303.31,
-    "train_steps_per_second": 8210.281
+    "train_loss": 0.21942725667586693,
+    "train_runtime": 100.7972,
+    "train_samples_per_second": 51.291,
+    "train_steps_per_second": 6.449
 }

config.json

@@ -28,7 +28,6 @@
   "num_channels": 3,
   "num_hidden_layers": 12,
   "patch_size": 16,
-  "problem_type": "single_label_classification",
   "qkv_bias": true,
   "torch_dtype": "float32",
   "transformers_version": "4.45.0.dev0"

eval_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 5.0,
     "eval_accuracy": 0.9924812030075187,
-    "eval_loss": 0.06499230116605759,
-    "eval_runtime": 1.0883,
-    "eval_samples_per_second": 122.212,
-    "eval_steps_per_second": 15.621
+    "eval_loss": 0.06558172404766083,
+    "eval_runtime": 1.0482,
+    "eval_samples_per_second": 126.888,
+    "eval_steps_per_second": 16.219
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6ab43b90e9a0dba3663bb520f27f1b133faa119571c5b3441c604f369343ff38
+oid sha256:9642438303d2b063dc124d1182a3b2a785cd9f704a965ed7052fc72f1c60318d
 size 343227052

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 5.0,
     "total_flos": 4.006371770595533e+17,
-    "train_loss": 0.0,
-    "train_runtime": 0.0792,
-    "train_samples_per_second": 65303.31,
-    "train_steps_per_second": 8210.281
+    "train_loss": 0.21942725667586693,
+    "train_runtime": 100.7972,
+    "train_samples_per_second": 51.291,
+    "train_steps_per_second": 6.449
 }

trainer_state.json

@@ -1,5 +1,5 @@
 {
-  "best_metric": 0.06499230116605759,
+  "best_metric": 0.06558172404766083,
   "best_model_checkpoint": "./beans_outputs/checkpoint-520",
   "epoch": 5.0,
   "eval_steps": 500,
@@ -10,512 +10,512 @@
   "log_history": [
     {
       "epoch": 0.07692307692307693,
-      "grad_norm": 2.198392391204834,
+      "grad_norm": 2.198366403579712,
       "learning_rate": 1.9692307692307696e-05,
       "loss": 1.0245,
       "step": 10
     },
     {
       "epoch": 0.15384615384615385,
-      "grad_norm": 1.846859335899353,
+      "grad_norm": 1.8460220098495483,
       "learning_rate": 1.9384615384615386e-05,
       "loss": 0.9453,
       "step": 20
     },
     {
       "epoch": 0.23076923076923078,
-      "grad_norm": 2.059973955154419,
+      "grad_norm": 2.117300271987915,
       "learning_rate": 1.907692307692308e-05,
-      "loss": 0.8405,
+      "loss": 0.8406,
       "step": 30
     },
     {
       "epoch": 0.3076923076923077,
-      "grad_norm": 2.618006944656372,
+      "grad_norm": 2.645448684692383,
       "learning_rate": 1.876923076923077e-05,
-      "loss": 0.6956,
+      "loss": 0.6951,
       "step": 40
     },
     {
       "epoch": 0.38461538461538464,
-      "grad_norm": 3.8514087200164795,
+      "grad_norm": 4.24210262298584,
       "learning_rate": 1.8461538461538465e-05,
-      "loss": 0.6615,
+      "loss": 0.662,
       "step": 50
     },
     {
       "epoch": 0.46153846153846156,
-      "grad_norm": 2.1362836360931396,
+      "grad_norm": 2.095567464828491,
       "learning_rate": 1.8153846153846155e-05,
-      "loss": 0.5829,
+      "loss": 0.5811,
       "step": 60
     },
     {
       "epoch": 0.5384615384615384,
-      "grad_norm": 1.7998223304748535,
+      "grad_norm": 2.079636573791504,
       "learning_rate": 1.784615384615385e-05,
-      "loss": 0.5202,
+      "loss": 0.5193,
       "step": 70
     },
     {
       "epoch": 0.6153846153846154,
-      "grad_norm": 1.2562752962112427,
+      "grad_norm": 1.218340277671814,
       "learning_rate": 1.753846153846154e-05,
-      "loss": 0.3936,
+      "loss": 0.3945,
       "step": 80
     },
     {
       "epoch": 0.6923076923076923,
-      "grad_norm": 2.561612606048584,
+      "grad_norm": 2.4416184425354004,
       "learning_rate": 1.7230769230769234e-05,
-      "loss": 0.3544,
+      "loss": 0.3528,
       "step": 90
     },
     {
       "epoch": 0.7692307692307693,
-      "grad_norm": 2.900808811187744,
+      "grad_norm": 2.8852720260620117,
       "learning_rate": 1.6923076923076924e-05,
-      "loss": 0.3681,
+      "loss": 0.3689,
       "step": 100
     },
     {
       "epoch": 0.8461538461538461,
-      "grad_norm": 2.010755777359009,
+      "grad_norm": 1.9503966569900513,
       "learning_rate": 1.6615384615384618e-05,
-      "loss": 0.3327,
+      "loss": 0.3368,
       "step": 110
     },
     {
       "epoch": 0.9230769230769231,
-      "grad_norm": 2.478074550628662,
+      "grad_norm": 2.335829496383667,
       "learning_rate": 1.630769230769231e-05,
-      "loss": 0.2444,
+      "loss": 0.2468,
       "step": 120
     },
     {
       "epoch": 1.0,
-      "grad_norm": 6.432145118713379,
+      "grad_norm": 7.0948638916015625,
       "learning_rate": 1.6000000000000003e-05,
-      "loss": 0.2815,
+      "loss": 0.2859,
       "step": 130
     },
     {
       "epoch": 1.0,
       "eval_accuracy": 0.9624060150375939,
-      "eval_loss": 0.21647371351718903,
-      "eval_runtime": 0.9083,
-      "eval_samples_per_second": 146.421,
-      "eval_steps_per_second": 18.715,
+      "eval_loss": 0.21892516314983368,
+      "eval_runtime": 0.9855,
+      "eval_samples_per_second": 134.957,
+      "eval_steps_per_second": 17.25,
       "step": 130
     },
     {
       "epoch": 1.0769230769230769,
-      "grad_norm": 1.6476765871047974,
+      "grad_norm": 1.765810251235962,
       "learning_rate": 1.5692307692307693e-05,
-      "loss": 0.1987,
+      "loss": 0.2006,
       "step": 140
     },
     {
       "epoch": 1.1538461538461537,
-      "grad_norm": 3.0852138996124268,
+      "grad_norm": 3.1403791904449463,
       "learning_rate": 1.5384615384615387e-05,
-      "loss": 0.2238,
+      "loss": 0.2243,
       "step": 150
     },
     {
       "epoch": 1.2307692307692308,
-      "grad_norm": 1.8425233364105225,
+      "grad_norm": 1.4933677911758423,
       "learning_rate": 1.5076923076923078e-05,
-      "loss": 0.2053,
+      "loss": 0.2035,
       "step": 160
     },
     {
       "epoch": 1.3076923076923077,
-      "grad_norm": 0.5158002972602844,
+      "grad_norm": 0.5171657800674438,
       "learning_rate": 1.4769230769230772e-05,
-      "loss": 0.2207,
+      "loss": 0.2208,
       "step": 170
     },
     {
       "epoch": 1.3846153846153846,
-      "grad_norm": 0.6001647710800171,
+      "grad_norm": 0.600874662399292,
       "learning_rate": 1.4461538461538462e-05,
-      "loss": 0.2113,
+      "loss": 0.211,
       "step": 180
     },
     {
       "epoch": 1.4615384615384617,
-      "grad_norm": 4.286410808563232,
+      "grad_norm": 3.5654282569885254,
       "learning_rate": 1.4153846153846156e-05,
-      "loss": 0.165,
+      "loss": 0.1607,
       "step": 190
     },
     {
       "epoch": 1.5384615384615383,
-      "grad_norm": 0.4559226334095001,
+      "grad_norm": 0.47061973810195923,
       "learning_rate": 1.3846153846153847e-05,
-      "loss": 0.2555,
+      "loss": 0.2526,
       "step": 200
     },
     {
       "epoch": 1.6153846153846154,
-      "grad_norm": 0.6161900162696838,
+      "grad_norm": 0.6342479586601257,
       "learning_rate": 1.353846153846154e-05,
-      "loss": 0.1819,
+      "loss": 0.1804,
       "step": 210
     },
     {
       "epoch": 1.6923076923076923,
-      "grad_norm": 2.301481008529663,
+      "grad_norm": 2.391136884689331,
       "learning_rate": 1.3230769230769231e-05,
-      "loss": 0.2104,
+      "loss": 0.2121,
       "step": 220
     },
     {
       "epoch": 1.7692307692307692,
-      "grad_norm": 1.8308448791503906,
+      "grad_norm": 1.8535972833633423,
       "learning_rate": 1.2923076923076925e-05,
-      "loss": 0.1762,
+      "loss": 0.1678,
       "step": 230
     },
     {
       "epoch": 1.8461538461538463,
-      "grad_norm": 5.278977394104004,
+      "grad_norm": 11.256393432617188,
       "learning_rate": 1.2615384615384616e-05,
-      "loss": 0.2573,
+      "loss": 0.2574,
       "step": 240
     },
     {
       "epoch": 1.9230769230769231,
-      "grad_norm": 0.40304750204086304,
+      "grad_norm": 0.4076194167137146,
       "learning_rate": 1.230769230769231e-05,
-      "loss": 0.1377,
+      "loss": 0.1442,
       "step": 250
     },
     {
       "epoch": 2.0,
-      "grad_norm": 0.350562185049057,
+      "grad_norm": 0.3490784168243408,
       "learning_rate": 1.2e-05,
-      "loss": 0.1287,
+      "loss": 0.1316,
       "step": 260
     },
     {
       "epoch": 2.0,
       "eval_accuracy": 0.9699248120300752,
-      "eval_loss": 0.13134175539016724,
-      "eval_runtime": 0.9453,
-      "eval_samples_per_second": 140.692,
-      "eval_steps_per_second": 17.983,
+      "eval_loss": 0.13337133824825287,
+      "eval_runtime": 0.9938,
+      "eval_samples_per_second": 133.827,
+      "eval_steps_per_second": 17.106,
       "step": 260
     },
     {
       "epoch": 2.076923076923077,
-      "grad_norm": 0.5022653341293335,
+      "grad_norm": 0.5328232645988464,
       "learning_rate": 1.1692307692307694e-05,
-      "loss": 0.0932,
+      "loss": 0.0897,
       "step": 270
     },
     {
       "epoch": 2.1538461538461537,
-      "grad_norm": 0.3385148346424103,
+      "grad_norm": 0.3375113904476166,
       "learning_rate": 1.1384615384615385e-05,
-      "loss": 0.145,
+      "loss": 0.1475,
       "step": 280
     },
     {
       "epoch": 2.230769230769231,
-      "grad_norm": 7.167346477508545,
+      "grad_norm": 7.195676326751709,
       "learning_rate": 1.1076923076923079e-05,
-      "loss": 0.1236,
+      "loss": 0.1321,
       "step": 290
     },
     {
       "epoch": 2.3076923076923075,
-      "grad_norm": 11.584583282470703,
+      "grad_norm": 8.226975440979004,
       "learning_rate": 1.076923076923077e-05,
-      "loss": 0.138,
+      "loss": 0.1375,
       "step": 300
     },
     {
       "epoch": 2.3846153846153846,
-      "grad_norm": 4.579661846160889,
+      "grad_norm": 5.051274299621582,
       "learning_rate": 1.0461538461538463e-05,
-      "loss": 0.1726,
+      "loss": 0.1755,
       "step": 310
     },
     {
       "epoch": 2.4615384615384617,
-      "grad_norm": 0.30623820424079895,
+      "grad_norm": 0.30559489130973816,
       "learning_rate": 1.0153846153846154e-05,
-      "loss": 0.1543,
+      "loss": 0.1542,
       "step": 320
     },
     {
       "epoch": 2.5384615384615383,
-      "grad_norm": 5.757983684539795,
+      "grad_norm": 6.551879405975342,
       "learning_rate": 9.846153846153848e-06,
-      "loss": 0.1897,
+      "loss": 0.1872,
       "step": 330
     },
     {
       "epoch": 2.6153846153846154,
-      "grad_norm": 1.116064190864563,
+      "grad_norm": 1.2321629524230957,
       "learning_rate": 9.53846153846154e-06,
-      "loss": 0.0727,
+      "loss": 0.0714,
       "step": 340
     },
     {
       "epoch": 2.6923076923076925,
-      "grad_norm": 0.3445896506309509,
+      "grad_norm": 0.3684569299221039,
       "learning_rate": 9.230769230769232e-06,
-      "loss": 0.1346,
+      "loss": 0.1325,
       "step": 350
     },
     {
       "epoch": 2.769230769230769,
-      "grad_norm": 12.3712158203125,
+      "grad_norm": 9.62742805480957,
       "learning_rate": 8.923076923076925e-06,
-      "loss": 0.1349,
+      "loss": 0.1203,
       "step": 360
     },
     {
       "epoch": 2.8461538461538463,
-      "grad_norm": 0.29760444164276123,
+      "grad_norm": 0.3335096836090088,
       "learning_rate": 8.615384615384617e-06,
-      "loss": 0.1519,
+      "loss": 0.154,
       "step": 370
     },
     {
       "epoch": 2.9230769230769234,
-      "grad_norm": 1.1171053647994995,
+      "grad_norm": 1.1392183303833008,
       "learning_rate": 8.307692307692309e-06,
-      "loss": 0.1278,
+      "loss": 0.1306,
       "step": 380
     },
     {
       "epoch": 3.0,
-      "grad_norm": 0.3179745078086853,
+      "grad_norm": 0.29587361216545105,
       "learning_rate": 8.000000000000001e-06,
-      "loss": 0.1408,
+      "loss": 0.1438,
       "step": 390
     },
     {
       "epoch": 3.0,
       "eval_accuracy": 0.9699248120300752,
-      "eval_loss": 0.09979864209890366,
-      "eval_runtime": 0.9058,
-      "eval_samples_per_second": 146.838,
-      "eval_steps_per_second": 18.769,
+      "eval_loss": 0.09808181971311569,
+      "eval_runtime": 0.9838,
+      "eval_samples_per_second": 135.192,
+      "eval_steps_per_second": 17.28,
       "step": 390
     },
     {
       "epoch": 3.076923076923077,
-      "grad_norm": 2.1718528270721436,
+      "grad_norm": 2.2563395500183105,
       "learning_rate": 7.692307692307694e-06,
-      "loss": 0.1321,
+      "loss": 0.1325,
       "step": 400
     },
     {
       "epoch": 3.1538461538461537,
-      "grad_norm": 6.615433216094971,
+      "grad_norm": 6.758065223693848,
       "learning_rate": 7.384615384615386e-06,
-      "loss": 0.1546,
+      "loss": 0.154,
       "step": 410
     },
     {
       "epoch": 3.230769230769231,
-      "grad_norm": 8.979787826538086,
+      "grad_norm": 8.712298393249512,
       "learning_rate": 7.076923076923078e-06,
-      "loss": 0.1429,
+      "loss": 0.1422,
       "step": 420
     },
     {
       "epoch": 3.3076923076923075,
-      "grad_norm": 0.2338067889213562,
+      "grad_norm": 0.23161759972572327,
       "learning_rate": 6.76923076923077e-06,
       "loss": 0.0768,
       "step": 430
     },
     {
       "epoch": 3.3846153846153846,
-      "grad_norm": 5.406207084655762,
+      "grad_norm": 5.164632797241211,
       "learning_rate": 6.461538461538463e-06,
-      "loss": 0.1158,
+      "loss": 0.1136,
       "step": 440
     },
     {
       "epoch": 3.4615384615384617,
-      "grad_norm": 12.034098625183105,
+      "grad_norm": 12.590032577514648,
       "learning_rate": 6.153846153846155e-06,
-      "loss": 0.0831,
+      "loss": 0.0854,
       "step": 450
     },
     {
       "epoch": 3.5384615384615383,
-      "grad_norm": 5.9186906814575195,
+      "grad_norm": 4.483827590942383,
       "learning_rate": 5.846153846153847e-06,
-      "loss": 0.0963,
+      "loss": 0.098,
       "step": 460
     },
     {
       "epoch": 3.6153846153846154,
-      "grad_norm": 8.250151634216309,
+      "grad_norm": 7.863719940185547,
       "learning_rate": 5.538461538461539e-06,
-      "loss": 0.0791,
+      "loss": 0.0778,
       "step": 470
     },
     {
       "epoch": 3.6923076923076925,
-      "grad_norm": 0.21918846666812897,
+      "grad_norm": 0.21892791986465454,
       "learning_rate": 5.230769230769232e-06,
-      "loss": 0.0971,
+      "loss": 0.0957,
       "step": 480
     },
     {
       "epoch": 3.769230769230769,
-      "grad_norm": 0.21208502352237701,
+      "grad_norm": 0.2107536494731903,
       "learning_rate": 4.923076923076924e-06,
-      "loss": 0.0718,
+      "loss": 0.0775,
       "step": 490
     },
     {
       "epoch": 3.8461538461538463,
-      "grad_norm": 0.9157105684280396,
+      "grad_norm": 0.9603075981140137,
       "learning_rate": 4.615384615384616e-06,
-      "loss": 0.1413,
+      "loss": 0.1357,
       "step": 500
     },
     {
       "epoch": 3.9230769230769234,
-      "grad_norm": 0.20776814222335815,
+      "grad_norm": 0.21214883029460907,
       "learning_rate": 4.307692307692308e-06,
-      "loss": 0.1272,
+      "loss": 0.1225,
       "step": 510
     },
     {
       "epoch": 4.0,
-      "grad_norm": 0.30272677540779114,
+      "grad_norm": 0.2996317744255066,
       "learning_rate": 4.000000000000001e-06,
-      "loss": 0.0846,
+      "loss": 0.0833,
       "step": 520
     },
     {
       "epoch": 4.0,
       "eval_accuracy": 0.9924812030075187,
-      "eval_loss": 0.06499230116605759,
-      "eval_runtime": 0.9407,
-      "eval_samples_per_second": 141.382,
-      "eval_steps_per_second": 18.071,
+      "eval_loss": 0.06558172404766083,
+      "eval_runtime": 1.0412,
+      "eval_samples_per_second": 127.735,
+      "eval_steps_per_second": 16.327,
       "step": 520
     },
     {
       "epoch": 4.076923076923077,
-      "grad_norm": 7.613215923309326,
+      "grad_norm": 9.374862670898438,
       "learning_rate": 3.692307692307693e-06,
-      "loss": 0.1002,
+      "loss": 0.1009,
       "step": 530
     },
     {
       "epoch": 4.153846153846154,
-      "grad_norm": 1.5558936595916748,
+      "grad_norm": 1.5623271465301514,
       "learning_rate": 3.384615384615385e-06,
-      "loss": 0.1657,
+      "loss": 0.1658,
       "step": 540
     },
     {
       "epoch": 4.230769230769231,
-      "grad_norm": 0.3433710038661957,
+      "grad_norm": 0.3211674094200134,
       "learning_rate": 3.0769230769230774e-06,
-      "loss": 0.1559,
+      "loss": 0.1534,
       "step": 550
     },
     {
       "epoch": 4.3076923076923075,
-      "grad_norm": 1.3209067583084106,
+      "grad_norm": 1.208674430847168,
       "learning_rate": 2.7692307692307697e-06,
-      "loss": 0.0526,
+      "loss": 0.0504,
       "step": 560
     },
     {
       "epoch": 4.384615384615385,
-      "grad_norm": 0.307217538356781,
+      "grad_norm": 0.3158506751060486,
       "learning_rate": 2.461538461538462e-06,
-      "loss": 0.1305,
+      "loss": 0.1345,
       "step": 570
     },
     {
       "epoch": 4.461538461538462,
-      "grad_norm": 0.273890882730484,
+      "grad_norm": 0.2690170705318451,
       "learning_rate": 2.153846153846154e-06,
-      "loss": 0.0813,
+      "loss": 0.0816,
       "step": 580
     },
     {
       "epoch": 4.538461538461538,
-      "grad_norm": 9.096426963806152,
+      "grad_norm": 9.395990371704102,
       "learning_rate": 1.8461538461538465e-06,
-      "loss": 0.0783,
+      "loss": 0.0714,
       "step": 590
     },
     {
       "epoch": 4.615384615384615,
-      "grad_norm": 3.9279935359954834,
+      "grad_norm": 3.067613124847412,
       "learning_rate": 1.5384615384615387e-06,
-      "loss": 0.0952,
+      "loss": 0.0892,
       "step": 600
     },
     {
       "epoch": 4.6923076923076925,
-      "grad_norm": 5.055220127105713,
+      "grad_norm": 5.167181491851807,
       "learning_rate": 1.230769230769231e-06,
-      "loss": 0.0868,
+      "loss": 0.087,
       "step": 610
     },
     {
       "epoch": 4.769230769230769,
-      "grad_norm": 0.2579015791416168,
+      "grad_norm": 0.2610948383808136,
       "learning_rate": 9.230769230769232e-07,
-      "loss": 0.0834,
+      "loss": 0.0807,
       "step": 620
     },
     {
       "epoch": 4.846153846153846,
-      "grad_norm": 0.2216470092535019,
+      "grad_norm": 0.22187209129333496,
       "learning_rate": 6.153846153846155e-07,
-      "loss": 0.0644,
+      "loss": 0.0668,
       "step": 630
     },
     {
       "epoch": 4.923076923076923,
-      "grad_norm": 0.23264001309871674,
+      "grad_norm": 0.23086942732334137,
       "learning_rate": 3.0769230769230774e-07,
-      "loss": 0.0775,
+      "loss": 0.0784,
       "step": 640
     },
     {
       "epoch": 5.0,
-      "grad_norm": 0.5718781352043152,
+      "grad_norm": 0.6445406675338745,
       "learning_rate": 0.0,
-      "loss": 0.1105,
+      "loss": 0.1107,
       "step": 650
     },
     {
       "epoch": 5.0,
       "eval_accuracy": 0.9774436090225563,
-      "eval_loss": 0.08365403115749359,
-      "eval_runtime": 0.9587,
-      "eval_samples_per_second": 138.736,
-      "eval_steps_per_second": 17.733,
+      "eval_loss": 0.0816693976521492,
+      "eval_runtime": 1.9096,
+      "eval_samples_per_second": 69.648,
+      "eval_steps_per_second": 8.902,
       "step": 650
     },
     {
       "epoch": 5.0,
       "step": 650,
       "total_flos": 4.006371770595533e+17,
-      "train_loss": 0.0,
-      "train_runtime": 0.0792,
-      "train_samples_per_second": 65303.31,
-      "train_steps_per_second": 8210.281
+      "train_loss": 0.21942725667586693,
+      "train_runtime": 100.7972,
+      "train_samples_per_second": 51.291,
+      "train_steps_per_second": 6.449
     }
   ],
   "logging_steps": 10,

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ec3cacd6d0ca0d3620eb7e4ee8e753968a6cc4a74d80f67a584244205c41f5fd
+oid sha256:519c14c04b23c1ab6e31ee48861306975f6c15dd2fdb90ddcfbf4c73b20b066c
 size 5240