Training in progress, step 2000, checkpoint

last-checkpoint/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:927cc3a6954f0ffa37523365e836fc8d0e054ba9e9b62e4efeef208689ca1e69
+size 499730128

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5b67e41ed00d2988f2a953732e3cea60fbca1c8978b56b795619d682c8546e07
-size 1442699
+oid sha256:721aca8d970b3613badea97e9ab731e82baf534a74aed8a73e3cbd0c140c5b80
+size 812034507

last-checkpoint/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6d920d35377b6218b0a182ca529dc7f210a3398080e4578011b7878234683794
+size 14645

last-checkpoint/tokenizer_config.json

@@ -0,0 +1,15 @@
+{
+  "backend": "tokenizers",
+  "cls_token": "[CLS]",
+  "do_lower_case": true,
+  "is_local": false,
+  "local_files_only": false,
+  "mask_token": "[MASK]",
+  "model_max_length": 1000000000,
+  "pad_token": "[PAD]",
+  "sep_token": "[SEP]",
+  "strip_accents": null,
+  "tokenize_chinese_chars": true,
+  "tokenizer_class": "BertTokenizer",
+  "unk_token": "[UNK]"
+}

last-checkpoint/trainer_state.json

@@ -6,165 +6,165 @@
   "eval_steps": 1000,
   "global_step": 2000,
   "is_hyper_param_search": false,
-  "is_local_process_zero": false,
+  "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
       "epoch": 0.002,
-      "grad_norm": 17.168848037719727,
+      "grad_norm": 1.2400550842285156,
       "learning_rate": 7.920000000000001e-05,
-      "loss": 67.9955,
+      "loss": 8.362694702148438,
       "step": 100
     },
     {
       "epoch": 0.004,
-      "grad_norm": 17.03850555419922,
+      "grad_norm": 2.1569149494171143,
       "learning_rate": 0.00015920000000000002,
-      "loss": 54.2312,
+      "loss": 6.592593383789063,
       "step": 200
     },
     {
       "epoch": 0.006,
-      "grad_norm": 14.042399406433105,
+      "grad_norm": 4.070979595184326,
       "learning_rate": 0.00023920000000000001,
-      "loss": 51.6192,
+      "loss": 6.208985595703125,
       "step": 300
     },
     {
       "epoch": 0.008,
-      "grad_norm": 27.294328689575195,
+      "grad_norm": 4.392622470855713,
       "learning_rate": 0.0003192,
-      "loss": 50.0348,
+      "loss": 5.921519775390625,
       "step": 400
     },
     {
       "epoch": 0.01,
-      "grad_norm": 24.557180404663086,
+      "grad_norm": 4.653286457061768,
       "learning_rate": 0.0003992,
-      "loss": 48.6163,
+      "loss": 5.658856201171875,
       "step": 500
     },
     {
       "epoch": 0.012,
-      "grad_norm": 16.080244064331055,
+      "grad_norm": 2.6610119342803955,
       "learning_rate": 0.00047920000000000005,
-      "loss": 47.19,
+      "loss": 5.450134887695312,
       "step": 600
     },
     {
       "epoch": 0.014,
-      "grad_norm": 11.565583229064941,
+      "grad_norm": 2.6298203468322754,
       "learning_rate": 0.0005592,
-      "loss": 46.5522,
+      "loss": 5.260469360351562,
       "step": 700
     },
     {
       "epoch": 0.016,
-      "grad_norm": 20.977285385131836,
+      "grad_norm": 1.681970238685608,
       "learning_rate": 0.0006392,
-      "loss": 46.0124,
+      "loss": 5.112535095214843,
       "step": 800
     },
     {
       "epoch": 0.018,
-      "grad_norm": 27.219560623168945,
+      "grad_norm": 1.5673184394836426,
       "learning_rate": 0.0007191999999999999,
-      "loss": 45.5052,
+      "loss": 5.0076907348632815,
       "step": 900
     },
     {
       "epoch": 0.02,
-      "grad_norm": 15.375182151794434,
+      "grad_norm": 1.6654455661773682,
       "learning_rate": 0.0007992,
-      "loss": 44.7058,
+      "loss": 4.9159228515625,
       "step": 1000
     },
     {
       "epoch": 0.02,
-      "eval_accuracy": 0.18614285714285714,
-      "eval_loss": 44.36890411376953,
-      "eval_runtime": 14.8109,
-      "eval_samples_per_second": 8.44,
-      "eval_steps_per_second": 0.135,
+      "eval_accuracy": 0.2274637964774951,
+      "eval_loss": 4.8702616691589355,
+      "eval_runtime": 5.1375,
+      "eval_samples_per_second": 194.646,
+      "eval_steps_per_second": 1.557,
       "step": 1000
     },
     {
       "epoch": 0.022,
-      "grad_norm": 14.803359031677246,
+      "grad_norm": 1.3562606573104858,
       "learning_rate": 0.0008792,
-      "loss": 44.1071,
+      "loss": 4.811459045410157,
       "step": 1100
     },
     {
       "epoch": 0.024,
-      "grad_norm": 9.71738338470459,
+      "grad_norm": 1.1230770349502563,
       "learning_rate": 0.0009592000000000001,
-      "loss": 43.516,
+      "loss": 4.759424438476563,
       "step": 1200
     },
     {
       "epoch": 0.026,
-      "grad_norm": 14.85153865814209,
+      "grad_norm": 0.9026045203208923,
       "learning_rate": 0.0010391999999999999,
-      "loss": 42.9438,
+      "loss": 4.691336669921875,
       "step": 1300
     },
     {
       "epoch": 0.028,
-      "grad_norm": 7.095604419708252,
+      "grad_norm": 0.8914104104042053,
       "learning_rate": 0.0011192,
-      "loss": 42.2365,
+      "loss": 4.6415283203125,
       "step": 1400
     },
     {
       "epoch": 0.03,
-      "grad_norm": 5.978736400604248,
+      "grad_norm": 0.8752658367156982,
       "learning_rate": 0.0011992,
-      "loss": 41.832,
+      "loss": 4.560531311035156,
       "step": 1500
     },
     {
       "epoch": 0.032,
-      "grad_norm": 7.7359161376953125,
+      "grad_norm": 0.7435582280158997,
       "learning_rate": 0.0012791999999999999,
-      "loss": 42.4716,
+      "loss": 4.516304016113281,
       "step": 1600
     },
     {
       "epoch": 0.034,
-      "grad_norm": 6.965914726257324,
+      "grad_norm": 0.7041499018669128,
       "learning_rate": 0.0013592,
-      "loss": 41.8079,
+      "loss": 4.48115234375,
       "step": 1700
     },
     {
       "epoch": 0.036,
-      "grad_norm": 7.237981796264648,
+      "grad_norm": 0.5327283143997192,
       "learning_rate": 0.0014392,
-      "loss": 41.2717,
+      "loss": 4.429443969726562,
       "step": 1800
     },
     {
       "epoch": 0.038,
-      "grad_norm": 6.8691253662109375,
+      "grad_norm": 0.45207348465919495,
       "learning_rate": 0.0015192,
-      "loss": 40.9756,
+      "loss": 4.37362060546875,
       "step": 1900
     },
     {
       "epoch": 0.04,
-      "grad_norm": 9.182296752929688,
+      "grad_norm": 0.5152547359466553,
       "learning_rate": 0.0015992,
-      "loss": 40.5064,
+      "loss": 4.377236938476562,
       "step": 2000
     },
     {
       "epoch": 0.04,
-      "eval_accuracy": 0.2189471624266145,
-      "eval_loss": 40.17017364501953,
-      "eval_runtime": 3.4954,
-      "eval_samples_per_second": 35.761,
-      "eval_steps_per_second": 0.572,
+      "eval_accuracy": 0.2694227005870842,
+      "eval_loss": 4.3582682609558105,
+      "eval_runtime": 5.3878,
+      "eval_samples_per_second": 185.606,
+      "eval_steps_per_second": 1.485,
       "step": 2000
     }
   ],
@@ -186,7 +186,7 @@
     }
   },
   "total_flos": 0.0,
-  "train_batch_size": 8,
+  "train_batch_size": 128,
   "trial_name": null,
   "trial_params": null
 }

last-checkpoint/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:13b7158ace9259ba4a46b6e596e8ae3278650c23f403ecbb51a56d4d4d36529b
-size 5905
+oid sha256:3ab34ea13b0e64af4c31a655731e444610b50d6a753b685ccf0d645cea5f76a3
+size 5265