Model save

README.md

@@ -1,6 +1,6 @@
 ---
 license: apache-2.0
-base_model: facebook/deit-base-distilled-patch16-224
+base_model: microsoft/beit-base-patch16-224-pt22k
 tags:
 - generated_from_trainer
 datasets:
@@ -22,7 +22,7 @@ model-index:
     metrics:
     - name: Accuracy
       type: accuracy
-      value: 0.9661654135338346
+      value: 0.9624060150375939
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -30,10 +30,10 @@ should probably proofread and complete it, then remove this comment. -->
 
 # finetuned-Leukemia-cell
 
-This model is a fine-tuned version of [facebook/deit-base-distilled-patch16-224](https://huggingface.co/facebook/deit-base-distilled-patch16-224) on the imagefolder dataset.
+This model is a fine-tuned version of [microsoft/beit-base-patch16-224-pt22k](https://huggingface.co/microsoft/beit-base-patch16-224-pt22k) on the imagefolder dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.0956
-- Accuracy: 0.9662
+- Loss: 0.0946
+- Accuracy: 0.9624
 
 ## Model description
 
@@ -65,12 +65,12 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch | Step | Validation Loss | Accuracy |
 |:-------------:|:-----:|:----:|:---------------:|:--------:|
-| 0.3464        | 2.94  | 100  | 0.2377          | 0.9286   |
-| 0.153         | 5.88  | 200  | 0.2519          | 0.9060   |
-| 0.1663        | 8.82  | 300  | 0.1753          | 0.9398   |
-| 0.0962        | 11.76 | 400  | 0.2009          | 0.9398   |
-| 0.0612        | 14.71 | 500  | 0.1348          | 0.9586   |
-| 0.0341        | 17.65 | 600  | 0.0956          | 0.9662   |
+| 0.9733        | 2.94  | 100  | 0.8894          | 0.7256   |
+| 0.7184        | 5.88  | 200  | 0.7876          | 0.7293   |
+| 0.5299        | 8.82  | 300  | 0.5183          | 0.8609   |
+| 0.3991        | 11.76 | 400  | 0.3121          | 0.8947   |
+| 0.2263        | 14.71 | 500  | 0.1337          | 0.9549   |
+| 0.1782        | 17.65 | 600  | 0.0946          | 0.9624   |
 
 
 ### Framework versions

all_results.json

@@ -1,13 +1,13 @@
 {
-    "epoch": 10.0,
-    "eval_accuracy": 0.9774436090225563,
-    "eval_loss": 0.08701933920383453,
-    "eval_runtime": 6.074,
-    "eval_samples_per_second": 43.794,
-    "eval_steps_per_second": 5.598,
-    "total_flos": 8.230018095787622e+17,
-    "train_loss": 0.1465345554492053,
-    "train_runtime": 221.0307,
-    "train_samples_per_second": 48.048,
-    "train_steps_per_second": 1.538
+    "epoch": 20.0,
+    "eval_accuracy": 0.9661654135338346,
+    "eval_loss": 0.09556835144758224,
+    "eval_runtime": 3.1465,
+    "eval_samples_per_second": 84.539,
+    "eval_steps_per_second": 10.806,
+    "total_flos": 1.646033084688384e+18,
+    "train_loss": 0.19354178275474732,
+    "train_runtime": 444.6785,
+    "train_samples_per_second": 47.765,
+    "train_steps_per_second": 1.529
 }

config.json

@@ -1,10 +1,14 @@
 {
-  "_name_or_path": "facebook/deit-base-distilled-patch16-224",
+  "_name_or_path": "microsoft/beit-base-patch16-224-pt22k",
   "architectures": [
-    "DeiTForImageClassification"
+    "BeitForImageClassification"
   ],
   "attention_probs_dropout_prob": 0.0,
-  "encoder_stride": 16,
+  "auxiliary_channels": 256,
+  "auxiliary_concat_input": false,
+  "auxiliary_loss_weight": 0.4,
+  "auxiliary_num_convs": 1,
+  "drop_path_rate": 0.1,
   "hidden_act": "gelu",
   "hidden_dropout_prob": 0.0,
   "hidden_size": 768,
@@ -30,13 +34,33 @@
     "trico": "6"
   },
   "layer_norm_eps": 1e-12,
-  "model_type": "deit",
+  "layer_scale_init_value": 0.1,
+  "model_type": "beit",
   "num_attention_heads": 12,
   "num_channels": 3,
   "num_hidden_layers": 12,
+  "out_indices": [
+    3,
+    5,
+    7,
+    11
+  ],
   "patch_size": 16,
+  "pool_scales": [
+    1,
+    2,
+    3,
+    6
+  ],
   "problem_type": "single_label_classification",
-  "qkv_bias": true,
+  "semantic_loss_ignore_index": 255,
   "torch_dtype": "float32",
-  "transformers_version": "4.35.2"
+  "transformers_version": "4.35.2",
+  "use_absolute_position_embeddings": false,
+  "use_auxiliary_head": true,
+  "use_mask_token": true,
+  "use_mean_pooling": true,
+  "use_relative_position_bias": false,
+  "use_shared_relative_position_bias": true,
+  "vocab_size": 8192
 }

eval_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 10.0,
-    "eval_accuracy": 0.9774436090225563,
-    "eval_loss": 0.08701933920383453,
-    "eval_runtime": 6.074,
-    "eval_samples_per_second": 43.794,
-    "eval_steps_per_second": 5.598
+    "epoch": 20.0,
+    "eval_accuracy": 0.9661654135338346,
+    "eval_loss": 0.09556835144758224,
+    "eval_runtime": 3.1465,
+    "eval_samples_per_second": 84.539,
+    "eval_steps_per_second": 10.806
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6776e9b0b9e3c3eb96201cecbd325ce233c78bbcd9bb45487d90adf02d5aa37c
-size 343245796
+oid sha256:ae4a6b3a2562aa63e1abd081ac90bbba7f0bef4b41d3240d4d8e328a08ced51d
+size 342710540

preprocessor_config.json

@@ -3,26 +3,27 @@
     "height": 224,
     "width": 224
   },
-  "do_center_crop": true,
+  "do_center_crop": false,
   "do_normalize": true,
+  "do_reduce_labels": false,
   "do_rescale": true,
   "do_resize": true,
-  "feature_extractor_type": "DeiTFeatureExtractor",
+  "feature_extractor_type": "BeitFeatureExtractor",
   "image_mean": [
-    0.485,
-    0.456,
-    0.406
+    0.5,
+    0.5,
+    0.5
   ],
-  "image_processor_type": "DeiTFeatureExtractor",
+  "image_processor_type": "BeitImageProcessor",
   "image_std": [
-    0.229,
-    0.224,
-    0.225
+    0.5,
+    0.5,
+    0.5
   ],
-  "resample": 3,
+  "resample": 2,
   "rescale_factor": 0.00392156862745098,
   "size": {
-    "height": 256,
-    "width": 256
+    "height": 224,
+    "width": 224
   }
 }

runs/Dec01_13-00-42_610b2a9400b8/events.out.tfevents.1701436123.610b2a9400b8.3653.4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8dcda7f2804c809ef20157f7ba1b65ba909b365c9a571bcbba6a9c1770079a26
+size 411

runs/Dec01_13-11-56_610b2a9400b8/events.out.tfevents.1701436344.610b2a9400b8.3653.5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2baa2cec02fcb787ee9a6c9dadaaf45a1a7ec0f1e0b33a8d1d013e7496cba707
+size 17923

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 10.0,
-    "total_flos": 8.230018095787622e+17,
-    "train_loss": 0.1465345554492053,
-    "train_runtime": 221.0307,
-    "train_samples_per_second": 48.048,
-    "train_steps_per_second": 1.538
+    "epoch": 20.0,
+    "total_flos": 1.646033084688384e+18,
+    "train_loss": 0.19354178275474732,
+    "train_runtime": 444.6785,
+    "train_samples_per_second": 47.765,
+    "train_steps_per_second": 1.529
 }

trainer_state.json

@@ -1,259 +1,490 @@
 {
-  "best_metric": 0.08701933920383453,
-  "best_model_checkpoint": "finetuned-Leukemia-cell/checkpoint-300",
-  "epoch": 10.0,
+  "best_metric": 0.09556835144758224,
+  "best_model_checkpoint": "finetuned-Leukemia-cell/checkpoint-600",
+  "epoch": 20.0,
   "eval_steps": 100,
-  "global_step": 340,
+  "global_step": 680,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
       "epoch": 0.29,
-      "learning_rate": 0.00019411764705882354,
-      "loss": 0.5837,
+      "learning_rate": 0.00019764705882352942,
+      "loss": 1.7578,
       "step": 10
     },
     {
       "epoch": 0.59,
-      "learning_rate": 0.00018823529411764707,
-      "loss": 0.2956,
+      "learning_rate": 0.0001947058823529412,
+      "loss": 1.016,
       "step": 20
     },
     {
       "epoch": 0.88,
-      "learning_rate": 0.0001823529411764706,
-      "loss": 0.279,
+      "learning_rate": 0.00019176470588235295,
+      "loss": 0.9006,
       "step": 30
     },
     {
       "epoch": 1.18,
-      "learning_rate": 0.00017647058823529413,
-      "loss": 0.1566,
+      "learning_rate": 0.00018882352941176472,
+      "loss": 0.7129,
       "step": 40
     },
     {
       "epoch": 1.47,
-      "learning_rate": 0.00017058823529411766,
-      "loss": 0.1042,
+      "learning_rate": 0.00018588235294117648,
+      "loss": 0.3781,
       "step": 50
     },
     {
       "epoch": 1.76,
-      "learning_rate": 0.0001647058823529412,
-      "loss": 0.176,
+      "learning_rate": 0.00018294117647058825,
+      "loss": 0.4442,
       "step": 60
     },
     {
       "epoch": 2.06,
-      "learning_rate": 0.0001588235294117647,
-      "loss": 0.1231,
+      "learning_rate": 0.00018,
+      "loss": 0.3716,
       "step": 70
     },
     {
       "epoch": 2.35,
-      "learning_rate": 0.00015294117647058822,
-      "loss": 0.0952,
+      "learning_rate": 0.00017705882352941178,
+      "loss": 0.371,
       "step": 80
     },
     {
       "epoch": 2.65,
-      "learning_rate": 0.00014705882352941178,
-      "loss": 0.1003,
+      "learning_rate": 0.00017411764705882354,
+      "loss": 0.3183,
       "step": 90
     },
     {
       "epoch": 2.94,
-      "learning_rate": 0.0001411764705882353,
-      "loss": 0.1606,
+      "learning_rate": 0.0001711764705882353,
+      "loss": 0.3464,
       "step": 100
     },
     {
       "epoch": 2.94,
-      "eval_accuracy": 0.9511278195488722,
-      "eval_loss": 0.18294382095336914,
-      "eval_runtime": 2.8739,
-      "eval_samples_per_second": 92.556,
-      "eval_steps_per_second": 11.83,
+      "eval_accuracy": 0.9285714285714286,
+      "eval_loss": 0.2377004772424698,
+      "eval_runtime": 103.2565,
+      "eval_samples_per_second": 2.576,
+      "eval_steps_per_second": 0.329,
       "step": 100
     },
     {
       "epoch": 3.24,
-      "learning_rate": 0.00013529411764705884,
-      "loss": 0.2204,
+      "learning_rate": 0.00016823529411764707,
+      "loss": 0.2195,
       "step": 110
     },
     {
       "epoch": 3.53,
-      "learning_rate": 0.00012941176470588237,
-      "loss": 0.2845,
+      "learning_rate": 0.0001652941176470588,
+      "loss": 0.3354,
       "step": 120
     },
     {
       "epoch": 3.82,
-      "learning_rate": 0.0001235294117647059,
-      "loss": 0.2104,
+      "learning_rate": 0.0001623529411764706,
+      "loss": 0.3365,
       "step": 130
     },
     {
       "epoch": 4.12,
-      "learning_rate": 0.00011764705882352942,
-      "loss": 0.2029,
+      "learning_rate": 0.00015941176470588237,
+      "loss": 0.2152,
       "step": 140
     },
     {
       "epoch": 4.41,
-      "learning_rate": 0.00011176470588235294,
-      "loss": 0.1105,
+      "learning_rate": 0.00015647058823529413,
+      "loss": 0.3078,
       "step": 150
     },
     {
       "epoch": 4.71,
-      "learning_rate": 0.00010588235294117647,
-      "loss": 0.177,
+      "learning_rate": 0.0001535294117647059,
+      "loss": 0.2712,
       "step": 160
     },
     {
       "epoch": 5.0,
-      "learning_rate": 0.0001,
-      "loss": 0.1881,
+      "learning_rate": 0.00015058823529411766,
+      "loss": 0.133,
       "step": 170
     },
     {
       "epoch": 5.29,
-      "learning_rate": 9.411764705882353e-05,
-      "loss": 0.1445,
+      "learning_rate": 0.00014764705882352943,
+      "loss": 0.3753,
       "step": 180
     },
     {
       "epoch": 5.59,
-      "learning_rate": 8.823529411764706e-05,
-      "loss": 0.0891,
+      "learning_rate": 0.0001447058823529412,
+      "loss": 0.2242,
       "step": 190
     },
     {
       "epoch": 5.88,
-      "learning_rate": 8.23529411764706e-05,
-      "loss": 0.1895,
+      "learning_rate": 0.00014176470588235296,
+      "loss": 0.153,
       "step": 200
     },
     {
       "epoch": 5.88,
-      "eval_accuracy": 0.9661654135338346,
-      "eval_loss": 0.14408943057060242,
-      "eval_runtime": 2.3757,
-      "eval_samples_per_second": 111.968,
-      "eval_steps_per_second": 14.312,
+      "eval_accuracy": 0.9060150375939849,
+      "eval_loss": 0.2518657147884369,
+      "eval_runtime": 2.626,
+      "eval_samples_per_second": 101.294,
+      "eval_steps_per_second": 12.947,
       "step": 200
     },
     {
       "epoch": 6.18,
-      "learning_rate": 7.647058823529411e-05,
-      "loss": 0.1095,
+      "learning_rate": 0.00013882352941176472,
+      "loss": 0.2496,
       "step": 210
     },
     {
       "epoch": 6.47,
-      "learning_rate": 7.058823529411765e-05,
-      "loss": 0.1908,
+      "learning_rate": 0.00013588235294117649,
+      "loss": 0.2584,
       "step": 220
     },
     {
       "epoch": 6.76,
-      "learning_rate": 6.470588235294118e-05,
-      "loss": 0.1118,
+      "learning_rate": 0.00013294117647058822,
+      "loss": 0.1437,
       "step": 230
     },
     {
       "epoch": 7.06,
-      "learning_rate": 5.882352941176471e-05,
-      "loss": 0.0461,
+      "learning_rate": 0.00013000000000000002,
+      "loss": 0.1991,
       "step": 240
     },
     {
       "epoch": 7.35,
-      "learning_rate": 5.294117647058824e-05,
-      "loss": 0.1002,
+      "learning_rate": 0.00012705882352941175,
+      "loss": 0.1776,
       "step": 250
     },
     {
       "epoch": 7.65,
-      "learning_rate": 4.705882352941177e-05,
-      "loss": 0.082,
+      "learning_rate": 0.00012411764705882355,
+      "loss": 0.133,
       "step": 260
     },
     {
       "epoch": 7.94,
-      "learning_rate": 4.11764705882353e-05,
-      "loss": 0.0712,
+      "learning_rate": 0.0001211764705882353,
+      "loss": 0.1119,
       "step": 270
     },
     {
       "epoch": 8.24,
-      "learning_rate": 3.529411764705883e-05,
-      "loss": 0.0565,
+      "learning_rate": 0.00011823529411764706,
+      "loss": 0.087,
       "step": 280
     },
     {
       "epoch": 8.53,
-      "learning_rate": 2.9411764705882354e-05,
-      "loss": 0.0702,
+      "learning_rate": 0.00011529411764705881,
+      "loss": 0.0795,
       "step": 290
     },
     {
       "epoch": 8.82,
-      "learning_rate": 2.3529411764705884e-05,
-      "loss": 0.0423,
+      "learning_rate": 0.00011235294117647059,
+      "loss": 0.1663,
       "step": 300
     },
     {
       "epoch": 8.82,
-      "eval_accuracy": 0.9774436090225563,
-      "eval_loss": 0.08701933920383453,
-      "eval_runtime": 2.414,
-      "eval_samples_per_second": 110.19,
-      "eval_steps_per_second": 14.084,
+      "eval_accuracy": 0.9398496240601504,
+      "eval_loss": 0.1752895563840866,
+      "eval_runtime": 2.5567,
+      "eval_samples_per_second": 104.042,
+      "eval_steps_per_second": 13.299,
       "step": 300
     },
     {
       "epoch": 9.12,
-      "learning_rate": 1.7647058823529414e-05,
-      "loss": 0.0652,
+      "learning_rate": 0.00010941176470588237,
+      "loss": 0.1149,
       "step": 310
     },
     {
       "epoch": 9.41,
-      "learning_rate": 1.1764705882352942e-05,
-      "loss": 0.027,
+      "learning_rate": 0.00010647058823529412,
+      "loss": 0.1415,
       "step": 320
     },
     {
       "epoch": 9.71,
-      "learning_rate": 5.882352941176471e-06,
-      "loss": 0.0633,
+      "learning_rate": 0.0001035294117647059,
+      "loss": 0.0998,
       "step": 330
     },
     {
       "epoch": 10.0,
-      "learning_rate": 0.0,
-      "loss": 0.055,
+      "learning_rate": 0.00010058823529411765,
+      "loss": 0.1299,
       "step": 340
     },
     {
-      "epoch": 10.0,
-      "step": 340,
-      "total_flos": 8.230018095787622e+17,
-      "train_loss": 0.1465345554492053,
-      "train_runtime": 221.0307,
-      "train_samples_per_second": 48.048,
-      "train_steps_per_second": 1.538
+      "epoch": 10.29,
+      "learning_rate": 9.764705882352942e-05,
+      "loss": 0.1028,
+      "step": 350
+    },
+    {
+      "epoch": 10.59,
+      "learning_rate": 9.470588235294118e-05,
+      "loss": 0.0636,
+      "step": 360
+    },
+    {
+      "epoch": 10.88,
+      "learning_rate": 9.176470588235295e-05,
+      "loss": 0.1064,
+      "step": 370
+    },
+    {
+      "epoch": 11.18,
+      "learning_rate": 8.882352941176471e-05,
+      "loss": 0.1138,
+      "step": 380
+    },
+    {
+      "epoch": 11.47,
+      "learning_rate": 8.588235294117646e-05,
+      "loss": 0.0469,
+      "step": 390
+    },
+    {
+      "epoch": 11.76,
+      "learning_rate": 8.294117647058824e-05,
+      "loss": 0.0962,
+      "step": 400
+    },
+    {
+      "epoch": 11.76,
+      "eval_accuracy": 0.9398496240601504,
+      "eval_loss": 0.20087337493896484,
+      "eval_runtime": 2.1402,
+      "eval_samples_per_second": 124.286,
+      "eval_steps_per_second": 15.886,
+      "step": 400
+    },
+    {
+      "epoch": 12.06,
+      "learning_rate": 8e-05,
+      "loss": 0.1099,
+      "step": 410
+    },
+    {
+      "epoch": 12.35,
+      "learning_rate": 7.705882352941177e-05,
+      "loss": 0.0475,
+      "step": 420
+    },
+    {
+      "epoch": 12.65,
+      "learning_rate": 7.411764705882354e-05,
+      "loss": 0.0448,
+      "step": 430
+    },
+    {
+      "epoch": 12.94,
+      "learning_rate": 7.11764705882353e-05,
+      "loss": 0.0782,
+      "step": 440
+    },
+    {
+      "epoch": 13.24,
+      "learning_rate": 6.823529411764707e-05,
+      "loss": 0.0543,
+      "step": 450
+    },
+    {
+      "epoch": 13.53,
+      "learning_rate": 6.529411764705883e-05,
+      "loss": 0.0252,
+      "step": 460
+    },
+    {
+      "epoch": 13.82,
+      "learning_rate": 6.23529411764706e-05,
+      "loss": 0.0517,
+      "step": 470
+    },
+    {
+      "epoch": 14.12,
+      "learning_rate": 5.9411764705882355e-05,
+      "loss": 0.0179,
+      "step": 480
+    },
+    {
+      "epoch": 14.41,
+      "learning_rate": 5.647058823529412e-05,
+      "loss": 0.0661,
+      "step": 490
+    },
+    {
+      "epoch": 14.71,
+      "learning_rate": 5.3529411764705884e-05,
+      "loss": 0.0612,
+      "step": 500
+    },
+    {
+      "epoch": 14.71,
+      "eval_accuracy": 0.9586466165413534,
+      "eval_loss": 0.13481324911117554,
+      "eval_runtime": 2.1446,
+      "eval_samples_per_second": 124.03,
+      "eval_steps_per_second": 15.853,
+      "step": 500
+    },
+    {
+      "epoch": 15.0,
+      "learning_rate": 5.058823529411765e-05,
+      "loss": 0.0611,
+      "step": 510
+    },
+    {
+      "epoch": 15.29,
+      "learning_rate": 4.7647058823529414e-05,
+      "loss": 0.0208,
+      "step": 520
+    },
+    {
+      "epoch": 15.59,
+      "learning_rate": 4.470588235294118e-05,
+      "loss": 0.0838,
+      "step": 530
+    },
+    {
+      "epoch": 15.88,
+      "learning_rate": 4.1764705882352944e-05,
+      "loss": 0.0558,
+      "step": 540
+    },
+    {
+      "epoch": 16.18,
+      "learning_rate": 3.882352941176471e-05,
+      "loss": 0.0544,
+      "step": 550
+    },
+    {
+      "epoch": 16.47,
+      "learning_rate": 3.5882352941176474e-05,
+      "loss": 0.0729,
+      "step": 560
+    },
+    {
+      "epoch": 16.76,
+      "learning_rate": 3.294117647058824e-05,
+      "loss": 0.0799,
+      "step": 570
+    },
+    {
+      "epoch": 17.06,
+      "learning_rate": 3e-05,
+      "loss": 0.011,
+      "step": 580
+    },
+    {
+      "epoch": 17.35,
+      "learning_rate": 2.7058823529411766e-05,
+      "loss": 0.0749,
+      "step": 590
+    },
+    {
+      "epoch": 17.65,
+      "learning_rate": 2.411764705882353e-05,
+      "loss": 0.0341,
+      "step": 600
+    },
+    {
+      "epoch": 17.65,
+      "eval_accuracy": 0.9661654135338346,
+      "eval_loss": 0.09556835144758224,
+      "eval_runtime": 2.4514,
+      "eval_samples_per_second": 108.508,
+      "eval_steps_per_second": 13.869,
+      "step": 600
+    },
+    {
+      "epoch": 17.94,
+      "learning_rate": 2.1176470588235296e-05,
+      "loss": 0.0134,
+      "step": 610
+    },
+    {
+      "epoch": 18.24,
+      "learning_rate": 1.8235294117647057e-05,
+      "loss": 0.0339,
+      "step": 620
+    },
+    {
+      "epoch": 18.53,
+      "learning_rate": 1.5294117647058826e-05,
+      "loss": 0.0383,
+      "step": 630
+    },
+    {
+      "epoch": 18.82,
+      "learning_rate": 1.2352941176470589e-05,
+      "loss": 0.0209,
+      "step": 640
+    },
+    {
+      "epoch": 19.12,
+      "learning_rate": 9.411764705882354e-06,
+      "loss": 0.0418,
+      "step": 650
+    },
+    {
+      "epoch": 19.41,
+      "learning_rate": 6.470588235294119e-06,
+      "loss": 0.0431,
+      "step": 660
+    },
+    {
+      "epoch": 19.71,
+      "learning_rate": 3.5294117647058825e-06,
+      "loss": 0.0334,
+      "step": 670
+    },
+    {
+      "epoch": 20.0,
+      "learning_rate": 5.882352941176471e-07,
+      "loss": 0.0208,
+      "step": 680
+    },
+    {
+      "epoch": 20.0,
+      "step": 680,
+      "total_flos": 1.646033084688384e+18,
+      "train_loss": 0.19354178275474732,
+      "train_runtime": 444.6785,
+      "train_samples_per_second": 47.765,
+      "train_steps_per_second": 1.529
     }
   ],
   "logging_steps": 10,
-  "max_steps": 340,
-  "num_train_epochs": 10,
+  "max_steps": 680,
+  "num_train_epochs": 20,
   "save_steps": 100,
-  "total_flos": 8.230018095787622e+17,
+  "total_flos": 1.646033084688384e+18,
   "trial_name": null,
   "trial_params": null
 }

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:dacbff944cbe4f5a09b4fac9f1c809f6343455d2e3919516f15e7fd6ecdc5c74
+oid sha256:22a2c928be698f630b880bb20978029c7af153c01476125f0fdff83c1eb2efab
 size 4600