Frenz/modelsent_test

README.md

@@ -21,13 +21,13 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [albert/albert-base-v2](https://huggingface.co/albert/albert-base-v2) on the None dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.2374
-- Accuracy: 0.9255
-- F1: 0.9255
-- Precision: 0.9255
-- Recall: 0.9255
-- Accuracy Label Negative: 0.9268
-- Accuracy Label Positive: 0.9243
+- Loss: 0.2379
+- Accuracy: 0.9261
+- F1: 0.9261
+- Precision: 0.9261
+- Recall: 0.9261
+- Accuracy Label Negative: 0.9242
+- Accuracy Label Positive: 0.9278
 
 ## Model description
 
@@ -52,7 +52,7 @@ The following hyperparameters were used during training:
 - seed: 42
 - gradient_accumulation_steps: 2
 - total_train_batch_size: 32
-- optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
+- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
 - lr_scheduler_type: linear
 - lr_scheduler_warmup_steps: 500
 - num_epochs: 3
@@ -61,23 +61,23 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch  | Step | Validation Loss | Accuracy | F1     | Precision | Recall | Accuracy Label Negative | Accuracy Label Positive |
 |:-------------:|:------:|:----:|:---------------:|:--------:|:------:|:---------:|:------:|:-----------------------:|:-----------------------:|
-| 1.0862        | 0.2442 | 100  | 0.5049          | 0.7789   | 0.7786 | 0.7832    | 0.7789 | 0.8295                  | 0.7314                  |
-| 0.5271        | 0.4884 | 200  | 0.2858          | 0.9023   | 0.9023 | 0.9040    | 0.9023 | 0.9306                  | 0.8757                  |
-| 0.4448        | 0.7326 | 300  | 0.2418          | 0.9163   | 0.9163 | 0.9163    | 0.9163 | 0.9091                  | 0.9231                  |
-| 0.5582        | 0.9768 | 400  | 0.2191          | 0.9236   | 0.9237 | 0.9237    | 0.9236 | 0.9268                  | 0.9207                  |
-| 0.4054        | 1.2198 | 500  | 0.2682          | 0.9145   | 0.9144 | 0.9178    | 0.9145 | 0.9558                  | 0.8757                  |
-| 0.3433        | 1.4640 | 600  | 0.2552          | 0.9151   | 0.9150 | 0.9157    | 0.9151 | 0.8902                  | 0.9385                  |
-| 0.5589        | 1.7082 | 700  | 0.2087          | 0.9133   | 0.9131 | 0.9144    | 0.9133 | 0.8813                  | 0.9432                  |
-| 0.2343        | 1.9524 | 800  | 0.2110          | 0.9181   | 0.9181 | 0.9188    | 0.9181 | 0.8939                  | 0.9408                  |
-| 0.2403        | 2.1954 | 900  | 0.2314          | 0.9224   | 0.9224 | 0.9227    | 0.9224 | 0.9318                  | 0.9136                  |
-| 0.2643        | 2.4396 | 1000 | 0.2996          | 0.9041   | 0.9039 | 0.9102    | 0.9041 | 0.9621                  | 0.8497                  |
-| 0.1856        | 2.6838 | 1100 | 0.2395          | 0.9218   | 0.9218 | 0.9222    | 0.9218 | 0.9343                  | 0.9101                  |
-| 0.3018        | 2.9280 | 1200 | 0.2376          | 0.9261   | 0.9261 | 0.9261    | 0.9261 | 0.9242                  | 0.9278                  |
+| 0.5403        | 0.2442 | 100  | 0.5274          | 0.7611   | 0.7596 | 0.7728    | 0.7611 | 0.8535                  | 0.6746                  |
+| 0.2673        | 0.4884 | 200  | 0.2806          | 0.8980   | 0.8980 | 0.8994    | 0.8980 | 0.9230                  | 0.8746                  |
+| 0.247         | 0.7326 | 300  | 0.2610          | 0.9029   | 0.9024 | 0.9074    | 0.9029 | 0.8434                  | 0.9586                  |
+| 0.2357        | 0.9768 | 400  | 0.2560          | 0.9084   | 0.9084 | 0.9096    | 0.9084 | 0.9318                  | 0.8864                  |
+| 0.2094        | 1.2198 | 500  | 0.3127          | 0.9090   | 0.9089 | 0.9123    | 0.9090 | 0.9508                  | 0.8698                  |
+| 0.1695        | 1.4640 | 600  | 0.2298          | 0.9188   | 0.9187 | 0.9189    | 0.9188 | 0.9053                  | 0.9314                  |
+| 0.2024        | 1.7082 | 700  | 0.2218          | 0.9206   | 0.9206 | 0.9214    | 0.9206 | 0.9394                  | 0.9030                  |
+| 0.1155        | 1.9524 | 800  | 0.2061          | 0.9236   | 0.9236 | 0.9236    | 0.9236 | 0.9192                  | 0.9278                  |
+| 0.1361        | 2.1954 | 900  | 0.2299          | 0.9218   | 0.9218 | 0.9226    | 0.9218 | 0.9407                  | 0.9041                  |
+| 0.1235        | 2.4396 | 1000 | 0.2668          | 0.9212   | 0.9212 | 0.9246    | 0.9212 | 0.9634                  | 0.8817                  |
+| 0.084         | 2.6838 | 1100 | 0.2733          | 0.9218   | 0.9218 | 0.9240    | 0.9218 | 0.9545                  | 0.8911                  |
+| 0.1326        | 2.9280 | 1200 | 0.2395          | 0.9249   | 0.9249 | 0.9249    | 0.9249 | 0.9192                  | 0.9302                  |
 
 
 ### Framework versions
 
-- Transformers 4.47.1
-- Pytorch 2.5.1+cu121
-- Datasets 3.2.0
+- Transformers 4.49.0
+- Pytorch 2.5.1+cu124
+- Datasets 3.3.2
 - Tokenizers 0.21.0

config.json

@@ -36,7 +36,7 @@
   "position_embedding_type": "absolute",
   "problem_type": "single_label_classification",
   "torch_dtype": "float32",
-  "transformers_version": "4.47.1",
+  "transformers_version": "4.49.0",
   "type_vocab_size": 2,
   "vocab_size": 30000
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:53bc6e9afc36255a6f3245c393a03dc0544a202734f36f5493f5dc6955e177ad
+oid sha256:efb9d7af05ede4873b6bd337d522ea7d4b84698e444489e42d9469fcf780586e
 size 46743912

trainer_state.json

@@ -10,1047 +10,1047 @@
   "log_history": [
     {
       "epoch": 0.02442002442002442,
-      "grad_norm": 43.265899658203125,
+      "grad_norm": 25.41930389404297,
       "learning_rate": 4.0000000000000003e-07,
-      "loss": 1.5045,
+      "loss": 0.8076,
       "step": 10
     },
     {
       "epoch": 0.04884004884004884,
-      "grad_norm": 37.62464141845703,
+      "grad_norm": 11.460826873779297,
       "learning_rate": 8.000000000000001e-07,
-      "loss": 1.4611,
+      "loss": 0.7516,
       "step": 20
     },
     {
       "epoch": 0.07326007326007326,
-      "grad_norm": 34.83750534057617,
+      "grad_norm": 7.831686973571777,
       "learning_rate": 1.2000000000000002e-06,
-      "loss": 1.3954,
+      "loss": 0.7034,
       "step": 30
     },
     {
       "epoch": 0.09768009768009768,
-      "grad_norm": 28.517704010009766,
+      "grad_norm": 7.234014987945557,
       "learning_rate": 1.6000000000000001e-06,
-      "loss": 1.3351,
+      "loss": 0.6996,
       "step": 40
     },
     {
       "epoch": 0.1221001221001221,
-      "grad_norm": 28.224279403686523,
+      "grad_norm": 5.096914768218994,
       "learning_rate": 2.0000000000000003e-06,
-      "loss": 1.3337,
+      "loss": 0.6809,
       "step": 50
     },
     {
       "epoch": 0.14652014652014653,
-      "grad_norm": 33.45441818237305,
+      "grad_norm": 8.638614654541016,
       "learning_rate": 2.4000000000000003e-06,
-      "loss": 1.3357,
+      "loss": 0.6892,
       "step": 60
     },
     {
       "epoch": 0.17094017094017094,
-      "grad_norm": 28.07544708251953,
+      "grad_norm": 13.534109115600586,
       "learning_rate": 2.8000000000000003e-06,
-      "loss": 1.2805,
+      "loss": 0.6603,
       "step": 70
     },
     {
       "epoch": 0.19536019536019536,
-      "grad_norm": 28.925960540771484,
+      "grad_norm": 12.838939666748047,
       "learning_rate": 3.2000000000000003e-06,
-      "loss": 1.2157,
+      "loss": 0.6225,
       "step": 80
     },
     {
       "epoch": 0.21978021978021978,
-      "grad_norm": 29.959428787231445,
+      "grad_norm": 14.190698623657227,
       "learning_rate": 3.6000000000000003e-06,
-      "loss": 1.1112,
+      "loss": 0.5716,
       "step": 90
     },
     {
       "epoch": 0.2442002442002442,
-      "grad_norm": 33.78961181640625,
+      "grad_norm": 18.23443031311035,
       "learning_rate": 4.000000000000001e-06,
-      "loss": 1.0862,
+      "loss": 0.5403,
       "step": 100
     },
     {
       "epoch": 0.2442002442002442,
-      "eval_accuracy": 0.7788637751985339,
-      "eval_accuracy_label_negative": 0.8295454545454546,
-      "eval_accuracy_label_positive": 0.7313609467455622,
-      "eval_f1": 0.7785675230312261,
-      "eval_loss": 0.5048791766166687,
-      "eval_precision": 0.7832195543584349,
-      "eval_recall": 0.7788637751985339,
-      "eval_runtime": 6.4624,
-      "eval_samples_per_second": 253.312,
-      "eval_steps_per_second": 15.938,
+      "eval_accuracy": 0.7611484422724496,
+      "eval_accuracy_label_negative": 0.8535353535353535,
+      "eval_accuracy_label_positive": 0.6745562130177515,
+      "eval_f1": 0.7596398965081889,
+      "eval_loss": 0.5274035930633545,
+      "eval_precision": 0.772811008310922,
+      "eval_recall": 0.7611484422724496,
+      "eval_runtime": 5.88,
+      "eval_samples_per_second": 278.402,
+      "eval_steps_per_second": 17.517,
       "step": 100
     },
     {
       "epoch": 0.2686202686202686,
-      "grad_norm": 224.80953979492188,
+      "grad_norm": 14.831952095031738,
       "learning_rate": 4.4e-06,
-      "loss": 0.9569,
+      "loss": 0.5022,
       "step": 110
     },
     {
       "epoch": 0.29304029304029305,
-      "grad_norm": 29.302122116088867,
+      "grad_norm": 18.24405860900879,
       "learning_rate": 4.800000000000001e-06,
-      "loss": 0.9296,
+      "loss": 0.4786,
       "step": 120
     },
     {
       "epoch": 0.31746031746031744,
-      "grad_norm": 50.676483154296875,
+      "grad_norm": 16.866863250732422,
       "learning_rate": 5.2e-06,
-      "loss": 0.7184,
+      "loss": 0.3629,
       "step": 130
     },
     {
       "epoch": 0.3418803418803419,
-      "grad_norm": 58.53715896606445,
+      "grad_norm": 18.194087982177734,
       "learning_rate": 5.600000000000001e-06,
-      "loss": 0.6639,
+      "loss": 0.3523,
       "step": 140
     },
     {
       "epoch": 0.3663003663003663,
-      "grad_norm": 46.839027404785156,
+      "grad_norm": 16.094165802001953,
       "learning_rate": 6e-06,
-      "loss": 0.7018,
+      "loss": 0.3272,
       "step": 150
     },
     {
       "epoch": 0.3907203907203907,
-      "grad_norm": 65.95608520507812,
+      "grad_norm": 41.393802642822266,
       "learning_rate": 6.4000000000000006e-06,
-      "loss": 0.5796,
+      "loss": 0.307,
       "step": 160
     },
     {
       "epoch": 0.41514041514041516,
-      "grad_norm": 51.919952392578125,
+      "grad_norm": 23.695520401000977,
       "learning_rate": 6.800000000000001e-06,
-      "loss": 0.6001,
+      "loss": 0.2551,
       "step": 170
     },
     {
       "epoch": 0.43956043956043955,
-      "grad_norm": 55.32732009887695,
+      "grad_norm": 30.489437103271484,
       "learning_rate": 7.2000000000000005e-06,
-      "loss": 0.5641,
+      "loss": 0.2536,
       "step": 180
     },
     {
       "epoch": 0.463980463980464,
-      "grad_norm": 74.93069458007812,
+      "grad_norm": 47.37874221801758,
       "learning_rate": 7.600000000000001e-06,
-      "loss": 0.5436,
+      "loss": 0.2622,
       "step": 190
     },
     {
       "epoch": 0.4884004884004884,
-      "grad_norm": 25.58298110961914,
+      "grad_norm": 26.136856079101562,
       "learning_rate": 8.000000000000001e-06,
-      "loss": 0.5271,
+      "loss": 0.2673,
       "step": 200
     },
     {
       "epoch": 0.4884004884004884,
-      "eval_accuracy": 0.9022602321319487,
-      "eval_accuracy_label_negative": 0.9305555555555556,
-      "eval_accuracy_label_positive": 0.8757396449704142,
-      "eval_f1": 0.9022657031312846,
-      "eval_loss": 0.2857775390148163,
-      "eval_precision": 0.9039560384223848,
-      "eval_recall": 0.9022602321319487,
-      "eval_runtime": 6.6059,
-      "eval_samples_per_second": 247.808,
-      "eval_steps_per_second": 15.592,
+      "eval_accuracy": 0.8979841172877214,
+      "eval_accuracy_label_negative": 0.922979797979798,
+      "eval_accuracy_label_positive": 0.8745562130177514,
+      "eval_f1": 0.8979978224223449,
+      "eval_loss": 0.2805691063404083,
+      "eval_precision": 0.8993693797991343,
+      "eval_recall": 0.8979841172877214,
+      "eval_runtime": 6.3687,
+      "eval_samples_per_second": 257.038,
+      "eval_steps_per_second": 16.173,
       "step": 200
     },
     {
       "epoch": 0.5128205128205128,
-      "grad_norm": 46.91861343383789,
+      "grad_norm": 17.713157653808594,
       "learning_rate": 8.400000000000001e-06,
-      "loss": 0.4353,
+      "loss": 0.1874,
       "step": 210
     },
     {
       "epoch": 0.5372405372405372,
-      "grad_norm": 84.31887817382812,
+      "grad_norm": 5.685952186584473,
       "learning_rate": 8.8e-06,
-      "loss": 0.4125,
+      "loss": 0.2565,
       "step": 220
     },
     {
       "epoch": 0.5616605616605617,
-      "grad_norm": 12.006002426147461,
+      "grad_norm": 64.32303619384766,
       "learning_rate": 9.200000000000002e-06,
-      "loss": 0.5151,
+      "loss": 0.2763,
       "step": 230
     },
     {
       "epoch": 0.5860805860805861,
-      "grad_norm": 9.51772403717041,
+      "grad_norm": 26.14358139038086,
       "learning_rate": 9.600000000000001e-06,
-      "loss": 0.5596,
+      "loss": 0.2193,
       "step": 240
     },
     {
       "epoch": 0.6105006105006106,
-      "grad_norm": 49.31715393066406,
+      "grad_norm": 51.215972900390625,
       "learning_rate": 1e-05,
-      "loss": 0.5192,
+      "loss": 0.2645,
       "step": 250
     },
     {
       "epoch": 0.6349206349206349,
-      "grad_norm": 99.71334838867188,
+      "grad_norm": 77.36116027832031,
       "learning_rate": 1.04e-05,
-      "loss": 0.4906,
+      "loss": 0.2887,
       "step": 260
     },
     {
       "epoch": 0.6593406593406593,
-      "grad_norm": 18.97690773010254,
+      "grad_norm": 12.419713020324707,
       "learning_rate": 1.0800000000000002e-05,
-      "loss": 0.5602,
+      "loss": 0.2836,
       "step": 270
     },
     {
       "epoch": 0.6837606837606838,
-      "grad_norm": 51.922142028808594,
+      "grad_norm": 51.27354049682617,
       "learning_rate": 1.1200000000000001e-05,
-      "loss": 0.4599,
+      "loss": 0.2279,
       "step": 280
     },
     {
       "epoch": 0.7081807081807082,
-      "grad_norm": 26.68296241760254,
+      "grad_norm": 15.182161331176758,
       "learning_rate": 1.16e-05,
-      "loss": 0.4887,
+      "loss": 0.23,
       "step": 290
     },
     {
       "epoch": 0.7326007326007326,
-      "grad_norm": 42.840782165527344,
+      "grad_norm": 20.174087524414062,
       "learning_rate": 1.2e-05,
-      "loss": 0.4448,
+      "loss": 0.247,
       "step": 300
     },
     {
       "epoch": 0.7326007326007326,
-      "eval_accuracy": 0.916310323762981,
-      "eval_accuracy_label_negative": 0.9090909090909091,
-      "eval_accuracy_label_positive": 0.9230769230769231,
-      "eval_f1": 0.9162971894414865,
-      "eval_loss": 0.24179790914058685,
-      "eval_precision": 0.916317612377141,
-      "eval_recall": 0.916310323762981,
-      "eval_runtime": 6.1349,
-      "eval_samples_per_second": 266.833,
-      "eval_steps_per_second": 16.789,
+      "eval_accuracy": 0.902871105681124,
+      "eval_accuracy_label_negative": 0.8434343434343434,
+      "eval_accuracy_label_positive": 0.9585798816568047,
+      "eval_f1": 0.9024095653522399,
+      "eval_loss": 0.26104938983917236,
+      "eval_precision": 0.9073823065773275,
+      "eval_recall": 0.902871105681124,
+      "eval_runtime": 6.46,
+      "eval_samples_per_second": 253.406,
+      "eval_steps_per_second": 15.944,
       "step": 300
     },
     {
       "epoch": 0.757020757020757,
-      "grad_norm": 42.57475280761719,
+      "grad_norm": 19.257923126220703,
       "learning_rate": 1.2400000000000002e-05,
-      "loss": 0.5235,
+      "loss": 0.2514,
       "step": 310
     },
     {
       "epoch": 0.7814407814407814,
-      "grad_norm": 15.34103775024414,
+      "grad_norm": 13.050935745239258,
       "learning_rate": 1.2800000000000001e-05,
-      "loss": 0.3713,
+      "loss": 0.216,
       "step": 320
     },
     {
       "epoch": 0.8058608058608059,
-      "grad_norm": 5.413728713989258,
+      "grad_norm": 1.8762781620025635,
       "learning_rate": 1.3200000000000002e-05,
-      "loss": 0.5713,
+      "loss": 0.2239,
       "step": 330
     },
     {
       "epoch": 0.8302808302808303,
-      "grad_norm": 13.065258026123047,
+      "grad_norm": 14.866827964782715,
       "learning_rate": 1.3600000000000002e-05,
-      "loss": 0.5092,
+      "loss": 0.2124,
       "step": 340
     },
     {
       "epoch": 0.8547008547008547,
-      "grad_norm": 29.859405517578125,
+      "grad_norm": 16.646745681762695,
       "learning_rate": 1.4e-05,
-      "loss": 0.5218,
+      "loss": 0.2341,
       "step": 350
     },
     {
       "epoch": 0.8791208791208791,
-      "grad_norm": 76.65414428710938,
+      "grad_norm": 14.468786239624023,
       "learning_rate": 1.4400000000000001e-05,
-      "loss": 0.5731,
+      "loss": 0.2397,
       "step": 360
     },
     {
       "epoch": 0.9035409035409036,
-      "grad_norm": 38.61219024658203,
+      "grad_norm": 16.370431900024414,
       "learning_rate": 1.48e-05,
-      "loss": 0.5184,
+      "loss": 0.2764,
       "step": 370
     },
     {
       "epoch": 0.927960927960928,
-      "grad_norm": 30.961179733276367,
+      "grad_norm": 15.854681015014648,
       "learning_rate": 1.5200000000000002e-05,
-      "loss": 0.3836,
+      "loss": 0.1964,
       "step": 380
     },
     {
       "epoch": 0.9523809523809523,
-      "grad_norm": 162.6136016845703,
+      "grad_norm": 19.42245101928711,
       "learning_rate": 1.5600000000000003e-05,
-      "loss": 0.4617,
+      "loss": 0.168,
       "step": 390
     },
     {
       "epoch": 0.9768009768009768,
-      "grad_norm": 14.536166191101074,
+      "grad_norm": 4.485373497009277,
       "learning_rate": 1.6000000000000003e-05,
-      "loss": 0.5582,
+      "loss": 0.2357,
       "step": 400
     },
     {
       "epoch": 0.9768009768009768,
-      "eval_accuracy": 0.9236408063530849,
-      "eval_accuracy_label_negative": 0.9267676767676768,
-      "eval_accuracy_label_positive": 0.9207100591715977,
-      "eval_f1": 0.9236520984092678,
-      "eval_loss": 0.21909849345684052,
-      "eval_precision": 0.9237192468386612,
-      "eval_recall": 0.9236408063530849,
-      "eval_runtime": 6.071,
-      "eval_samples_per_second": 269.644,
-      "eval_steps_per_second": 16.966,
+      "eval_accuracy": 0.9083689676237019,
+      "eval_accuracy_label_negative": 0.9318181818181818,
+      "eval_accuracy_label_positive": 0.8863905325443787,
+      "eval_f1": 0.9083847657945654,
+      "eval_loss": 0.2560252249240875,
+      "eval_precision": 0.9095968991483547,
+      "eval_recall": 0.9083689676237019,
+      "eval_runtime": 6.4906,
+      "eval_samples_per_second": 252.21,
+      "eval_steps_per_second": 15.869,
       "step": 400
     },
     {
       "epoch": 1.0,
-      "grad_norm": 8.102620124816895,
+      "grad_norm": 112.14476013183594,
       "learning_rate": 1.64e-05,
-      "loss": 0.3272,
+      "loss": 0.2322,
       "step": 410
     },
     {
       "epoch": 1.0244200244200243,
-      "grad_norm": 102.95278930664062,
+      "grad_norm": 19.950773239135742,
       "learning_rate": 1.6800000000000002e-05,
-      "loss": 0.354,
+      "loss": 0.1607,
       "step": 420
     },
     {
       "epoch": 1.0488400488400489,
-      "grad_norm": 19.004304885864258,
+      "grad_norm": 14.963955879211426,
       "learning_rate": 1.72e-05,
-      "loss": 0.4163,
+      "loss": 0.1943,
       "step": 430
     },
     {
       "epoch": 1.0732600732600732,
-      "grad_norm": 3.0345256328582764,
+      "grad_norm": 12.724336624145508,
       "learning_rate": 1.76e-05,
-      "loss": 0.2336,
+      "loss": 0.1273,
       "step": 440
     },
     {
       "epoch": 1.0976800976800978,
-      "grad_norm": 6.5663604736328125,
+      "grad_norm": 12.087287902832031,
       "learning_rate": 1.8e-05,
-      "loss": 0.4468,
+      "loss": 0.2392,
       "step": 450
     },
     {
       "epoch": 1.122100122100122,
-      "grad_norm": 15.335294723510742,
+      "grad_norm": 4.449314117431641,
       "learning_rate": 1.8400000000000003e-05,
-      "loss": 0.5378,
+      "loss": 0.2444,
       "step": 460
     },
     {
       "epoch": 1.1465201465201464,
-      "grad_norm": 23.567869186401367,
+      "grad_norm": 13.28513240814209,
       "learning_rate": 1.88e-05,
-      "loss": 0.4598,
+      "loss": 0.1995,
       "step": 470
     },
     {
       "epoch": 1.170940170940171,
-      "grad_norm": 119.62120819091797,
+      "grad_norm": 9.664812088012695,
       "learning_rate": 1.9200000000000003e-05,
-      "loss": 0.4447,
+      "loss": 0.2178,
       "step": 480
     },
     {
       "epoch": 1.1953601953601953,
-      "grad_norm": 5.290420055389404,
+      "grad_norm": 3.9961328506469727,
       "learning_rate": 1.9600000000000002e-05,
-      "loss": 0.214,
+      "loss": 0.1269,
       "step": 490
     },
     {
       "epoch": 1.2197802197802199,
-      "grad_norm": 220.08334350585938,
+      "grad_norm": 6.78753137588501,
       "learning_rate": 2e-05,
-      "loss": 0.4054,
+      "loss": 0.2094,
       "step": 500
     },
     {
       "epoch": 1.2197802197802199,
-      "eval_accuracy": 0.9144777031154551,
-      "eval_accuracy_label_negative": 0.9558080808080808,
-      "eval_accuracy_label_positive": 0.8757396449704142,
-      "eval_f1": 0.9144397212931268,
-      "eval_loss": 0.26820144057273865,
-      "eval_precision": 0.9177552790543617,
-      "eval_recall": 0.9144777031154551,
-      "eval_runtime": 6.1069,
-      "eval_samples_per_second": 268.058,
-      "eval_steps_per_second": 16.866,
+      "eval_accuracy": 0.9089798411728772,
+      "eval_accuracy_label_negative": 0.9507575757575758,
+      "eval_accuracy_label_positive": 0.8698224852071006,
+      "eval_f1": 0.9089364277932478,
+      "eval_loss": 0.31265538930892944,
+      "eval_precision": 0.9123227286547785,
+      "eval_recall": 0.9089798411728772,
+      "eval_runtime": 6.4966,
+      "eval_samples_per_second": 251.979,
+      "eval_steps_per_second": 15.855,
       "step": 500
     },
     {
       "epoch": 1.2442002442002442,
-      "grad_norm": 25.83207893371582,
+      "grad_norm": 4.199306488037109,
       "learning_rate": 1.9724896836313617e-05,
-      "loss": 0.461,
+      "loss": 0.1993,
       "step": 510
     },
     {
       "epoch": 1.2686202686202686,
-      "grad_norm": 33.92952346801758,
+      "grad_norm": 47.193603515625,
       "learning_rate": 1.9449793672627236e-05,
-      "loss": 0.4965,
+      "loss": 0.1837,
       "step": 520
     },
     {
       "epoch": 1.293040293040293,
-      "grad_norm": 82.4642105102539,
+      "grad_norm": 22.490142822265625,
       "learning_rate": 1.9174690508940855e-05,
-      "loss": 0.5272,
+      "loss": 0.2295,
       "step": 530
     },
     {
       "epoch": 1.3174603174603174,
-      "grad_norm": 15.041220664978027,
+      "grad_norm": 10.449468612670898,
       "learning_rate": 1.889958734525447e-05,
-      "loss": 0.5795,
+      "loss": 0.2589,
       "step": 540
     },
     {
       "epoch": 1.341880341880342,
-      "grad_norm": 51.05344009399414,
+      "grad_norm": 49.228878021240234,
       "learning_rate": 1.862448418156809e-05,
-      "loss": 0.3848,
+      "loss": 0.134,
       "step": 550
     },
     {
       "epoch": 1.3663003663003663,
-      "grad_norm": 363.78424072265625,
+      "grad_norm": 43.437068939208984,
       "learning_rate": 1.8349381017881708e-05,
-      "loss": 0.5898,
+      "loss": 0.2905,
       "step": 560
     },
     {
       "epoch": 1.3907203907203907,
-      "grad_norm": 118.15901947021484,
+      "grad_norm": 22.296707153320312,
       "learning_rate": 1.8074277854195327e-05,
-      "loss": 0.4777,
+      "loss": 0.2304,
       "step": 570
     },
     {
       "epoch": 1.4151404151404152,
-      "grad_norm": 20.32166862487793,
+      "grad_norm": 13.2725248336792,
       "learning_rate": 1.7799174690508942e-05,
-      "loss": 0.4722,
+      "loss": 0.2818,
       "step": 580
     },
     {
       "epoch": 1.4395604395604396,
-      "grad_norm": 12.633702278137207,
+      "grad_norm": 4.746452808380127,
       "learning_rate": 1.7524071526822558e-05,
-      "loss": 0.3581,
+      "loss": 0.1635,
       "step": 590
     },
     {
       "epoch": 1.463980463980464,
-      "grad_norm": 29.648452758789062,
+      "grad_norm": 15.717751502990723,
       "learning_rate": 1.7248968363136177e-05,
-      "loss": 0.3433,
+      "loss": 0.1695,
       "step": 600
     },
     {
       "epoch": 1.463980463980464,
-      "eval_accuracy": 0.9150885766646304,
-      "eval_accuracy_label_negative": 0.8901515151515151,
-      "eval_accuracy_label_positive": 0.9384615384615385,
-      "eval_f1": 0.9149907006403145,
-      "eval_loss": 0.25520163774490356,
-      "eval_precision": 0.91573365916924,
-      "eval_recall": 0.9150885766646304,
-      "eval_runtime": 6.0849,
-      "eval_samples_per_second": 269.025,
-      "eval_steps_per_second": 16.927,
+      "eval_accuracy": 0.9187538179596824,
+      "eval_accuracy_label_negative": 0.9053030303030303,
+      "eval_accuracy_label_positive": 0.9313609467455621,
+      "eval_f1": 0.9187176730282378,
+      "eval_loss": 0.22982099652290344,
+      "eval_precision": 0.9188801834239158,
+      "eval_recall": 0.9187538179596824,
+      "eval_runtime": 6.5054,
+      "eval_samples_per_second": 251.638,
+      "eval_steps_per_second": 15.833,
       "step": 600
     },
     {
       "epoch": 1.4884004884004884,
-      "grad_norm": 48.567466735839844,
+      "grad_norm": 25.554302215576172,
       "learning_rate": 1.6973865199449796e-05,
-      "loss": 0.3987,
+      "loss": 0.1926,
       "step": 610
     },
     {
       "epoch": 1.5128205128205128,
-      "grad_norm": 17.67637825012207,
+      "grad_norm": 6.077470779418945,
       "learning_rate": 1.669876203576341e-05,
-      "loss": 0.4218,
+      "loss": 0.2561,
       "step": 620
     },
     {
       "epoch": 1.537240537240537,
-      "grad_norm": 40.53282928466797,
+      "grad_norm": 19.69429588317871,
       "learning_rate": 1.642365887207703e-05,
-      "loss": 0.386,
+      "loss": 0.1501,
       "step": 630
     },
     {
       "epoch": 1.5616605616605617,
-      "grad_norm": 116.74990844726562,
+      "grad_norm": 10.95654010772705,
       "learning_rate": 1.614855570839065e-05,
-      "loss": 0.4431,
+      "loss": 0.1852,
       "step": 640
     },
     {
       "epoch": 1.5860805860805862,
-      "grad_norm": 20.529993057250977,
+      "grad_norm": 24.821308135986328,
       "learning_rate": 1.5873452544704268e-05,
-      "loss": 0.4856,
+      "loss": 0.2113,
       "step": 650
     },
     {
       "epoch": 1.6105006105006106,
-      "grad_norm": 16.94083595275879,
+      "grad_norm": 13.17242431640625,
       "learning_rate": 1.5598349381017883e-05,
-      "loss": 0.2793,
+      "loss": 0.1237,
       "step": 660
     },
     {
       "epoch": 1.6349206349206349,
-      "grad_norm": 43.13039779663086,
+      "grad_norm": 16.253307342529297,
       "learning_rate": 1.53232462173315e-05,
-      "loss": 0.338,
+      "loss": 0.1646,
       "step": 670
     },
     {
       "epoch": 1.6593406593406592,
-      "grad_norm": 56.50369644165039,
+      "grad_norm": 28.406679153442383,
       "learning_rate": 1.5048143053645119e-05,
-      "loss": 0.4709,
+      "loss": 0.2448,
       "step": 680
     },
     {
       "epoch": 1.6837606837606838,
-      "grad_norm": 13.253159523010254,
+      "grad_norm": 15.36701488494873,
       "learning_rate": 1.4773039889958736e-05,
-      "loss": 0.3916,
+      "loss": 0.1902,
       "step": 690
     },
     {
       "epoch": 1.7081807081807083,
-      "grad_norm": 25.500383377075195,
+      "grad_norm": 33.92890548706055,
       "learning_rate": 1.4497936726272352e-05,
-      "loss": 0.5589,
+      "loss": 0.2024,
       "step": 700
     },
     {
       "epoch": 1.7081807081807083,
-      "eval_accuracy": 0.9132559560171044,
-      "eval_accuracy_label_negative": 0.8813131313131313,
-      "eval_accuracy_label_positive": 0.9431952662721893,
-      "eval_f1": 0.9131080020430126,
-      "eval_loss": 0.20865951478481293,
-      "eval_precision": 0.9144124548952469,
-      "eval_recall": 0.9132559560171044,
-      "eval_runtime": 6.1693,
-      "eval_samples_per_second": 265.344,
-      "eval_steps_per_second": 16.695,
+      "eval_accuracy": 0.9205864386072083,
+      "eval_accuracy_label_negative": 0.9393939393939394,
+      "eval_accuracy_label_positive": 0.9029585798816568,
+      "eval_f1": 0.920605585060701,
+      "eval_loss": 0.22175996005535126,
+      "eval_precision": 0.9214190422768549,
+      "eval_recall": 0.9205864386072083,
+      "eval_runtime": 7.338,
+      "eval_samples_per_second": 223.086,
+      "eval_steps_per_second": 14.037,
       "step": 700
     },
     {
       "epoch": 1.7326007326007327,
-      "grad_norm": 26.2749080657959,
+      "grad_norm": 39.820068359375,
       "learning_rate": 1.4222833562585972e-05,
-      "loss": 0.3424,
+      "loss": 0.1628,
       "step": 710
     },
     {
       "epoch": 1.757020757020757,
-      "grad_norm": 24.71459197998047,
+      "grad_norm": 14.987146377563477,
       "learning_rate": 1.3947730398899588e-05,
-      "loss": 0.3582,
+      "loss": 0.1733,
       "step": 720
     },
     {
       "epoch": 1.7814407814407813,
-      "grad_norm": 16.336734771728516,
+      "grad_norm": 7.142777442932129,
       "learning_rate": 1.3672627235213205e-05,
-      "loss": 0.3334,
+      "loss": 0.1429,
       "step": 730
     },
     {
       "epoch": 1.8058608058608059,
-      "grad_norm": 11.974506378173828,
+      "grad_norm": 40.524967193603516,
       "learning_rate": 1.3397524071526824e-05,
-      "loss": 0.3375,
+      "loss": 0.1887,
       "step": 740
     },
     {
       "epoch": 1.8302808302808304,
-      "grad_norm": 35.09166717529297,
+      "grad_norm": 15.210890769958496,
       "learning_rate": 1.3122420907840441e-05,
-      "loss": 0.4735,
+      "loss": 0.2109,
       "step": 750
     },
     {
       "epoch": 1.8547008547008548,
-      "grad_norm": 23.616647720336914,
+      "grad_norm": 14.5702543258667,
       "learning_rate": 1.284731774415406e-05,
-      "loss": 0.3001,
+      "loss": 0.1397,
       "step": 760
     },
     {
       "epoch": 1.879120879120879,
-      "grad_norm": 31.474788665771484,
+      "grad_norm": 20.027212142944336,
       "learning_rate": 1.2572214580467677e-05,
-      "loss": 0.2949,
+      "loss": 0.1263,
       "step": 770
     },
     {
       "epoch": 1.9035409035409034,
-      "grad_norm": 41.20528793334961,
+      "grad_norm": 13.660886764526367,
       "learning_rate": 1.2297111416781292e-05,
-      "loss": 0.5207,
+      "loss": 0.2006,
       "step": 780
     },
     {
       "epoch": 1.927960927960928,
-      "grad_norm": 12.699962615966797,
+      "grad_norm": 8.097176551818848,
       "learning_rate": 1.2022008253094913e-05,
-      "loss": 0.4461,
+      "loss": 0.1901,
       "step": 790
     },
     {
       "epoch": 1.9523809523809523,
-      "grad_norm": 23.412885665893555,
+      "grad_norm": 4.4079108238220215,
       "learning_rate": 1.1746905089408529e-05,
-      "loss": 0.2343,
+      "loss": 0.1155,
       "step": 800
     },
     {
       "epoch": 1.9523809523809523,
-      "eval_accuracy": 0.918142944410507,
-      "eval_accuracy_label_negative": 0.8939393939393939,
-      "eval_accuracy_label_positive": 0.9408284023668639,
-      "eval_f1": 0.9180523324913266,
-      "eval_loss": 0.21100710332393646,
-      "eval_precision": 0.918757729196307,
-      "eval_recall": 0.918142944410507,
-      "eval_runtime": 6.1463,
-      "eval_samples_per_second": 266.338,
-      "eval_steps_per_second": 16.758,
+      "eval_accuracy": 0.9236408063530849,
+      "eval_accuracy_label_negative": 0.9191919191919192,
+      "eval_accuracy_label_positive": 0.927810650887574,
+      "eval_f1": 0.923636013636066,
+      "eval_loss": 0.20613956451416016,
+      "eval_precision": 0.9236374318453521,
+      "eval_recall": 0.9236408063530849,
+      "eval_runtime": 6.4768,
+      "eval_samples_per_second": 252.75,
+      "eval_steps_per_second": 15.903,
       "step": 800
     },
     {
       "epoch": 1.9768009768009769,
-      "grad_norm": 23.129348754882812,
+      "grad_norm": 16.103919982910156,
       "learning_rate": 1.1471801925722146e-05,
-      "loss": 0.402,
+      "loss": 0.1695,
       "step": 810
     },
     {
       "epoch": 2.0,
-      "grad_norm": 82.90428161621094,
+      "grad_norm": 26.186500549316406,
       "learning_rate": 1.1196698762035765e-05,
-      "loss": 0.3556,
+      "loss": 0.1878,
       "step": 820
     },
     {
       "epoch": 2.0244200244200243,
-      "grad_norm": 28.78835678100586,
+      "grad_norm": 8.27334213256836,
       "learning_rate": 1.0921595598349382e-05,
-      "loss": 0.2684,
+      "loss": 0.116,
       "step": 830
     },
     {
       "epoch": 2.0488400488400487,
-      "grad_norm": 7.678134918212891,
+      "grad_norm": 9.153972625732422,
       "learning_rate": 1.0646492434663e-05,
-      "loss": 0.2465,
+      "loss": 0.1331,
       "step": 840
     },
     {
       "epoch": 2.0732600732600734,
-      "grad_norm": 10.281868934631348,
+      "grad_norm": 8.734851837158203,
       "learning_rate": 1.0371389270976618e-05,
-      "loss": 0.2608,
+      "loss": 0.1064,
       "step": 850
     },
     {
       "epoch": 2.0976800976800978,
-      "grad_norm": 44.59360122680664,
+      "grad_norm": 16.807472229003906,
       "learning_rate": 1.0096286107290235e-05,
-      "loss": 0.2796,
+      "loss": 0.1323,
       "step": 860
     },
     {
       "epoch": 2.122100122100122,
-      "grad_norm": 71.7291259765625,
+      "grad_norm": 7.338029861450195,
       "learning_rate": 9.821182943603852e-06,
-      "loss": 0.1796,
+      "loss": 0.0876,
       "step": 870
     },
     {
       "epoch": 2.1465201465201464,
-      "grad_norm": 58.75422668457031,
+      "grad_norm": 51.00392150878906,
       "learning_rate": 9.54607977991747e-06,
-      "loss": 0.2029,
+      "loss": 0.0574,
       "step": 880
     },
     {
       "epoch": 2.1709401709401708,
-      "grad_norm": 24.300424575805664,
+      "grad_norm": 15.209708213806152,
       "learning_rate": 9.270976616231088e-06,
-      "loss": 0.4012,
+      "loss": 0.1536,
       "step": 890
     },
     {
       "epoch": 2.1953601953601956,
-      "grad_norm": 17.3212833404541,
+      "grad_norm": 5.431528568267822,
       "learning_rate": 8.995873452544705e-06,
-      "loss": 0.2403,
+      "loss": 0.1361,
       "step": 900
     },
     {
       "epoch": 2.1953601953601956,
-      "eval_accuracy": 0.9224190592547342,
-      "eval_accuracy_label_negative": 0.9318181818181818,
-      "eval_accuracy_label_positive": 0.9136094674556213,
-      "eval_f1": 0.9224377694184351,
-      "eval_loss": 0.23138785362243652,
-      "eval_precision": 0.9227050113183035,
-      "eval_recall": 0.9224190592547342,
-      "eval_runtime": 6.1138,
-      "eval_samples_per_second": 267.756,
-      "eval_steps_per_second": 16.847,
+      "eval_accuracy": 0.921808185705559,
+      "eval_accuracy_label_negative": 0.9406565656565656,
+      "eval_accuracy_label_positive": 0.9041420118343195,
+      "eval_f1": 0.9218270375982287,
+      "eval_loss": 0.2299249768257141,
+      "eval_precision": 0.9226412544476359,
+      "eval_recall": 0.921808185705559,
+      "eval_runtime": 6.4607,
+      "eval_samples_per_second": 253.378,
+      "eval_steps_per_second": 15.943,
       "step": 900
     },
     {
       "epoch": 2.21978021978022,
-      "grad_norm": 41.47114944458008,
+      "grad_norm": 18.443998336791992,
       "learning_rate": 8.720770288858322e-06,
-      "loss": 0.3843,
+      "loss": 0.1881,
       "step": 910
     },
     {
       "epoch": 2.244200244200244,
-      "grad_norm": 34.233089447021484,
+      "grad_norm": 18.574554443359375,
       "learning_rate": 8.44566712517194e-06,
-      "loss": 0.2175,
+      "loss": 0.0946,
       "step": 920
     },
     {
       "epoch": 2.2686202686202686,
-      "grad_norm": 63.74410629272461,
+      "grad_norm": 27.33018684387207,
       "learning_rate": 8.170563961485558e-06,
-      "loss": 0.3676,
+      "loss": 0.1302,
       "step": 930
     },
     {
       "epoch": 2.293040293040293,
-      "grad_norm": 52.46253204345703,
+      "grad_norm": 15.219032287597656,
       "learning_rate": 7.895460797799176e-06,
-      "loss": 0.3562,
+      "loss": 0.1475,
       "step": 940
     },
     {
       "epoch": 2.317460317460317,
-      "grad_norm": 113.09469604492188,
+      "grad_norm": 23.795124053955078,
       "learning_rate": 7.620357634112793e-06,
-      "loss": 0.2521,
+      "loss": 0.1098,
       "step": 950
     },
     {
       "epoch": 2.341880341880342,
-      "grad_norm": 49.29931640625,
+      "grad_norm": 12.746439933776855,
       "learning_rate": 7.345254470426411e-06,
-      "loss": 0.1803,
+      "loss": 0.0811,
       "step": 960
     },
     {
       "epoch": 2.3663003663003663,
-      "grad_norm": 35.607364654541016,
+      "grad_norm": 2.0957300662994385,
       "learning_rate": 7.070151306740029e-06,
-      "loss": 0.1935,
+      "loss": 0.1033,
       "step": 970
     },
     {
       "epoch": 2.3907203907203907,
-      "grad_norm": 59.298606872558594,
+      "grad_norm": 20.963695526123047,
       "learning_rate": 6.795048143053645e-06,
-      "loss": 0.3756,
+      "loss": 0.1716,
       "step": 980
     },
     {
       "epoch": 2.415140415140415,
-      "grad_norm": 24.38463592529297,
+      "grad_norm": 6.032730579376221,
       "learning_rate": 6.519944979367263e-06,
-      "loss": 0.2457,
+      "loss": 0.0937,
       "step": 990
     },
     {
       "epoch": 2.4395604395604398,
-      "grad_norm": 64.04911041259766,
+      "grad_norm": 18.961658477783203,
       "learning_rate": 6.244841815680881e-06,
-      "loss": 0.2643,
+      "loss": 0.1235,
       "step": 1000
     },
     {
       "epoch": 2.4395604395604398,
-      "eval_accuracy": 0.9040928527794746,
-      "eval_accuracy_label_negative": 0.9621212121212122,
-      "eval_accuracy_label_positive": 0.8497041420118343,
-      "eval_f1": 0.9039399934696009,
-      "eval_loss": 0.29960158467292786,
-      "eval_precision": 0.9101812880425119,
-      "eval_recall": 0.9040928527794746,
-      "eval_runtime": 6.1028,
-      "eval_samples_per_second": 268.238,
-      "eval_steps_per_second": 16.878,
+      "eval_accuracy": 0.9211973121563837,
+      "eval_accuracy_label_negative": 0.9633838383838383,
+      "eval_accuracy_label_positive": 0.8816568047337278,
+      "eval_f1": 0.9211597260760334,
+      "eval_loss": 0.26676130294799805,
+      "eval_precision": 0.9245979890029079,
+      "eval_recall": 0.9211973121563837,
+      "eval_runtime": 6.4755,
+      "eval_samples_per_second": 252.8,
+      "eval_steps_per_second": 15.906,
       "step": 1000
     },
     {
       "epoch": 2.463980463980464,
-      "grad_norm": 11.74494457244873,
+      "grad_norm": 11.030694961547852,
       "learning_rate": 5.969738651994499e-06,
-      "loss": 0.3761,
+      "loss": 0.1298,
       "step": 1010
     },
     {
       "epoch": 2.4884004884004884,
-      "grad_norm": 58.20482635498047,
+      "grad_norm": 19.397335052490234,
       "learning_rate": 5.6946354883081155e-06,
-      "loss": 0.2507,
+      "loss": 0.1046,
       "step": 1020
     },
     {
       "epoch": 2.5128205128205128,
-      "grad_norm": 8.351099967956543,
+      "grad_norm": 5.86328125,
       "learning_rate": 5.4195323246217335e-06,
-      "loss": 0.1995,
+      "loss": 0.0831,
       "step": 1030
     },
     {
       "epoch": 2.537240537240537,
-      "grad_norm": 52.496883392333984,
+      "grad_norm": 5.8682427406311035,
       "learning_rate": 5.1444291609353515e-06,
-      "loss": 0.3185,
+      "loss": 0.1152,
       "step": 1040
     },
     {
       "epoch": 2.5616605616605614,
-      "grad_norm": 15.573514938354492,
+      "grad_norm": 6.465247631072998,
       "learning_rate": 4.869325997248969e-06,
-      "loss": 0.3546,
+      "loss": 0.1316,
       "step": 1050
     },
     {
       "epoch": 2.586080586080586,
-      "grad_norm": 9.369500160217285,
+      "grad_norm": 74.92926025390625,
       "learning_rate": 4.594222833562587e-06,
-      "loss": 0.2971,
+      "loss": 0.0973,
       "step": 1060
     },
     {
       "epoch": 2.6105006105006106,
-      "grad_norm": 15.763602256774902,
+      "grad_norm": 4.780877590179443,
       "learning_rate": 4.319119669876204e-06,
-      "loss": 0.2744,
+      "loss": 0.1206,
       "step": 1070
     },
     {
       "epoch": 2.634920634920635,
-      "grad_norm": 43.78425979614258,
+      "grad_norm": 23.055227279663086,
       "learning_rate": 4.044016506189822e-06,
-      "loss": 0.2521,
+      "loss": 0.1001,
       "step": 1080
     },
     {
       "epoch": 2.659340659340659,
-      "grad_norm": 43.655250549316406,
+      "grad_norm": 50.33488082885742,
       "learning_rate": 3.768913342503439e-06,
-      "loss": 0.3189,
+      "loss": 0.1537,
       "step": 1090
     },
     {
       "epoch": 2.683760683760684,
-      "grad_norm": 21.811946868896484,
+      "grad_norm": 0.44596701860427856,
       "learning_rate": 3.4938101788170562e-06,
-      "loss": 0.1856,
+      "loss": 0.084,
       "step": 1100
     },
     {
       "epoch": 2.683760683760684,
       "eval_accuracy": 0.921808185705559,
-      "eval_accuracy_label_negative": 0.9343434343434344,
-      "eval_accuracy_label_positive": 0.9100591715976332,
-      "eval_f1": 0.9218285003622981,
-      "eval_loss": 0.23950175940990448,
-      "eval_precision": 0.922245080437082,
+      "eval_accuracy_label_negative": 0.9545454545454546,
+      "eval_accuracy_label_positive": 0.8911242603550296,
+      "eval_f1": 0.9218032836901542,
+      "eval_loss": 0.2732768654823303,
+      "eval_precision": 0.9239586674515976,
       "eval_recall": 0.921808185705559,
-      "eval_runtime": 6.1632,
-      "eval_samples_per_second": 265.61,
-      "eval_steps_per_second": 16.712,
+      "eval_runtime": 6.4795,
+      "eval_samples_per_second": 252.643,
+      "eval_steps_per_second": 15.896,
       "step": 1100
     },
     {
       "epoch": 2.7081807081807083,
-      "grad_norm": 15.966821670532227,
+      "grad_norm": 3.0834202766418457,
       "learning_rate": 3.2187070151306742e-06,
-      "loss": 0.2284,
+      "loss": 0.1546,
       "step": 1110
     },
     {
       "epoch": 2.7326007326007327,
-      "grad_norm": 34.287715911865234,
+      "grad_norm": 6.8805928230285645,
       "learning_rate": 2.943603851444292e-06,
-      "loss": 0.2117,
+      "loss": 0.0876,
       "step": 1120
     },
     {
       "epoch": 2.757020757020757,
-      "grad_norm": 45.06178665161133,
+      "grad_norm": 0.9852960109710693,
       "learning_rate": 2.6685006877579094e-06,
-      "loss": 0.204,
+      "loss": 0.0884,
       "step": 1130
     },
     {
       "epoch": 2.7814407814407813,
-      "grad_norm": 50.90498733520508,
+      "grad_norm": 9.610347747802734,
       "learning_rate": 2.393397524071527e-06,
-      "loss": 0.3576,
+      "loss": 0.1781,
       "step": 1140
     },
     {
       "epoch": 2.8058608058608057,
-      "grad_norm": 3.0487897396087646,
+      "grad_norm": 14.589119911193848,
       "learning_rate": 2.1182943603851446e-06,
-      "loss": 0.2927,
+      "loss": 0.1584,
       "step": 1150
     },
     {
       "epoch": 2.8302808302808304,
-      "grad_norm": 35.04338836669922,
+      "grad_norm": 5.850516319274902,
       "learning_rate": 1.8431911966987622e-06,
-      "loss": 0.3385,
+      "loss": 0.1266,
       "step": 1160
     },
     {
       "epoch": 2.8547008547008548,
-      "grad_norm": 2.87778902053833,
+      "grad_norm": 3.190080165863037,
       "learning_rate": 1.56808803301238e-06,
-      "loss": 0.3517,
+      "loss": 0.1563,
       "step": 1170
     },
     {
       "epoch": 2.879120879120879,
-      "grad_norm": 17.577260971069336,
+      "grad_norm": 8.95765209197998,
       "learning_rate": 1.2929848693259976e-06,
-      "loss": 0.1655,
+      "loss": 0.0896,
       "step": 1180
     },
     {
       "epoch": 2.9035409035409034,
-      "grad_norm": 5.3439106941223145,
+      "grad_norm": 1.606522798538208,
       "learning_rate": 1.017881705639615e-06,
-      "loss": 0.2477,
+      "loss": 0.1167,
       "step": 1190
     },
     {
       "epoch": 2.927960927960928,
-      "grad_norm": 56.782222747802734,
+      "grad_norm": 31.470670700073242,
       "learning_rate": 7.427785419532325e-07,
-      "loss": 0.3018,
+      "loss": 0.1326,
       "step": 1200
     },
     {
       "epoch": 2.927960927960928,
-      "eval_accuracy": 0.9260843005497862,
-      "eval_accuracy_label_negative": 0.9242424242424242,
-      "eval_accuracy_label_positive": 0.927810650887574,
-      "eval_f1": 0.9260857363073682,
-      "eval_loss": 0.23755837976932526,
-      "eval_precision": 0.9260878637932374,
-      "eval_recall": 0.9260843005497862,
-      "eval_runtime": 6.1684,
-      "eval_samples_per_second": 265.384,
-      "eval_steps_per_second": 16.698,
+      "eval_accuracy": 0.9248625534514355,
+      "eval_accuracy_label_negative": 0.9191919191919192,
+      "eval_accuracy_label_positive": 0.9301775147928995,
+      "eval_f1": 0.9248544119857385,
+      "eval_loss": 0.2394767850637436,
+      "eval_precision": 0.9248635488205493,
+      "eval_recall": 0.9248625534514355,
+      "eval_runtime": 6.4694,
+      "eval_samples_per_second": 253.036,
+      "eval_steps_per_second": 15.921,
       "step": 1200
     },
     {
       "epoch": 2.9523809523809526,
-      "grad_norm": 31.97544288635254,
+      "grad_norm": 16.254497528076172,
       "learning_rate": 4.6767537826685014e-07,
-      "loss": 0.2681,
+      "loss": 0.1154,
       "step": 1210
     },
     {
       "epoch": 2.976800976800977,
-      "grad_norm": 24.56979751586914,
+      "grad_norm": 18.331233978271484,
       "learning_rate": 1.925722145804677e-07,
-      "loss": 0.3463,
+      "loss": 0.1268,
       "step": 1220
     },
     {
       "epoch": 2.993894993894994,
       "step": 1227,
       "total_flos": 157098652813440.0,
-      "train_loss": 0.4734744669075603,
-      "train_runtime": 674.8973,
-      "train_samples_per_second": 58.187,
-      "train_steps_per_second": 1.818
+      "train_loss": 0.22647988077286502,
+      "train_runtime": 692.1682,
+      "train_samples_per_second": 56.735,
+      "train_steps_per_second": 1.773
     },
     {
       "epoch": 2.993894993894994,
-      "eval_accuracy": 0.9254734270006109,
-      "eval_accuracy_label_negative": 0.9267676767676768,
-      "eval_accuracy_label_positive": 0.9242603550295858,
-      "eval_f1": 0.9254812761220444,
-      "eval_loss": 0.23741039633750916,
-      "eval_precision": 0.9255140033647618,
-      "eval_recall": 0.9254734270006109,
-      "eval_runtime": 6.1438,
-      "eval_samples_per_second": 266.449,
-      "eval_steps_per_second": 16.765,
+      "eval_accuracy": 0.9260843005497862,
+      "eval_accuracy_label_negative": 0.9242424242424242,
+      "eval_accuracy_label_positive": 0.927810650887574,
+      "eval_f1": 0.9260857363073682,
+      "eval_loss": 0.23794083297252655,
+      "eval_precision": 0.9260878637932374,
+      "eval_recall": 0.9260843005497862,
+      "eval_runtime": 6.4878,
+      "eval_samples_per_second": 252.32,
+      "eval_steps_per_second": 15.876,
       "step": 1227
     }
   ],

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6329715e4875f017daad987374dcbe11cf4d98f0586fc36cf8b34f500c7c3287
+oid sha256:a337fbf44afe7056ea7d2ba6e5b5fcc3a1c2c2c6b87676da4257ff9d0586eb8c
 size 5304