Training in progress, step 2000, checkpoint

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:721aca8d970b3613badea97e9ab731e82baf534a74aed8a73e3cbd0c140c5b80
-size 812034507
+oid sha256:27f5349faf73282fae15b6197ce207ae54649a387b8a2ac9d9d6289a07744cb1
+size 1442699

last-checkpoint/pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:05fce6752ecbb6604e085ca1097f9b912b27b72a35695df4d7aa7a7834acef80
+oid sha256:a3234ebdb3c9bd5374463f6ff57f5856cedb2a1db06c4a225719dc1fe2d4e844
 size 499774435

last-checkpoint/trainer_state.json

@@ -6,165 +6,165 @@
   "eval_steps": 1000,
   "global_step": 2000,
   "is_hyper_param_search": false,
-  "is_local_process_zero": true,
+  "is_local_process_zero": false,
   "is_world_process_zero": true,
   "log_history": [
     {
       "epoch": 0.002,
-      "grad_norm": 1.2400550842285156,
+      "grad_norm": 19.660429000854492,
       "learning_rate": 7.920000000000001e-05,
-      "loss": 8.362694702148438,
+      "loss": 67.8543,
       "step": 100
     },
     {
       "epoch": 0.004,
-      "grad_norm": 2.1569149494171143,
+      "grad_norm": 24.802082061767578,
       "learning_rate": 0.00015920000000000002,
-      "loss": 6.592593383789063,
+      "loss": 54.1444,
       "step": 200
     },
     {
       "epoch": 0.006,
-      "grad_norm": 4.070979595184326,
+      "grad_norm": 16.059066772460938,
       "learning_rate": 0.00023920000000000001,
-      "loss": 6.208985595703125,
+      "loss": 51.5921,
       "step": 300
     },
     {
       "epoch": 0.008,
-      "grad_norm": 4.392622470855713,
+      "grad_norm": 27.705520629882812,
       "learning_rate": 0.0003192,
-      "loss": 5.921519775390625,
+      "loss": 50.0536,
       "step": 400
     },
     {
       "epoch": 0.01,
-      "grad_norm": 4.653286457061768,
+      "grad_norm": 25.79052734375,
       "learning_rate": 0.0003992,
-      "loss": 5.658856201171875,
+      "loss": 48.6987,
       "step": 500
     },
     {
       "epoch": 0.012,
-      "grad_norm": 2.6610119342803955,
+      "grad_norm": 14.82126522064209,
       "learning_rate": 0.00047920000000000005,
-      "loss": 5.450134887695312,
+      "loss": 47.3182,
       "step": 600
     },
     {
       "epoch": 0.014,
-      "grad_norm": 2.6298203468322754,
+      "grad_norm": 12.00988483428955,
       "learning_rate": 0.0005592,
-      "loss": 5.260469360351562,
+      "loss": 46.6574,
       "step": 700
     },
     {
       "epoch": 0.016,
-      "grad_norm": 1.681970238685608,
+      "grad_norm": 32.48664474487305,
       "learning_rate": 0.0006392,
-      "loss": 5.112535095214843,
+      "loss": 46.0747,
       "step": 800
     },
     {
       "epoch": 0.018,
-      "grad_norm": 1.5673184394836426,
+      "grad_norm": 25.646282196044922,
       "learning_rate": 0.0007191999999999999,
-      "loss": 5.0076907348632815,
+      "loss": 45.6098,
       "step": 900
     },
     {
       "epoch": 0.02,
-      "grad_norm": 1.6654455661773682,
+      "grad_norm": 16.071203231811523,
       "learning_rate": 0.0007992,
-      "loss": 4.9159228515625,
+      "loss": 44.8537,
       "step": 1000
     },
     {
       "epoch": 0.02,
-      "eval_accuracy": 0.2274637964774951,
-      "eval_loss": 4.8702616691589355,
-      "eval_runtime": 5.1375,
-      "eval_samples_per_second": 194.646,
-      "eval_steps_per_second": 1.557,
+      "eval_accuracy": 0.18606457925636008,
+      "eval_loss": 44.18140411376953,
+      "eval_runtime": 14.581,
+      "eval_samples_per_second": 8.573,
+      "eval_steps_per_second": 0.137,
       "step": 1000
     },
     {
       "epoch": 0.022,
-      "grad_norm": 1.3562606573104858,
+      "grad_norm": 14.277446746826172,
       "learning_rate": 0.0008792,
-      "loss": 4.811459045410157,
+      "loss": 44.223,
       "step": 1100
     },
     {
       "epoch": 0.024,
-      "grad_norm": 1.1230770349502563,
+      "grad_norm": 9.809870719909668,
       "learning_rate": 0.0009592000000000001,
-      "loss": 4.759424438476563,
+      "loss": 43.5719,
       "step": 1200
     },
     {
       "epoch": 0.026,
-      "grad_norm": 0.9026045203208923,
+      "grad_norm": 16.994014739990234,
       "learning_rate": 0.0010391999999999999,
-      "loss": 4.691336669921875,
+      "loss": 43.0323,
       "step": 1300
     },
     {
       "epoch": 0.028,
-      "grad_norm": 0.8914104104042053,
+      "grad_norm": 8.082005500793457,
       "learning_rate": 0.0011192,
-      "loss": 4.6415283203125,
+      "loss": 42.3631,
       "step": 1400
     },
     {
       "epoch": 0.03,
-      "grad_norm": 0.8752658367156982,
+      "grad_norm": 7.080636024475098,
       "learning_rate": 0.0011992,
-      "loss": 4.560531311035156,
+      "loss": 42.0285,
       "step": 1500
     },
     {
       "epoch": 0.032,
-      "grad_norm": 0.7435582280158997,
+      "grad_norm": 8.547815322875977,
       "learning_rate": 0.0012791999999999999,
-      "loss": 4.516304016113281,
+      "loss": 42.6829,
       "step": 1600
     },
     {
       "epoch": 0.034,
-      "grad_norm": 0.7041499018669128,
+      "grad_norm": 7.541939735412598,
       "learning_rate": 0.0013592,
-      "loss": 4.48115234375,
+      "loss": 41.9796,
       "step": 1700
     },
     {
       "epoch": 0.036,
-      "grad_norm": 0.5327283143997192,
+      "grad_norm": 6.902923583984375,
       "learning_rate": 0.0014392,
-      "loss": 4.429443969726562,
+      "loss": 41.5016,
       "step": 1800
     },
     {
       "epoch": 0.038,
-      "grad_norm": 0.45207348465919495,
+      "grad_norm": 7.68479061126709,
       "learning_rate": 0.0015192,
-      "loss": 4.37362060546875,
+      "loss": 41.2607,
       "step": 1900
     },
     {
       "epoch": 0.04,
-      "grad_norm": 0.5152547359466553,
+      "grad_norm": 8.936702728271484,
       "learning_rate": 0.0015992,
-      "loss": 4.377236938476562,
+      "loss": 40.7545,
       "step": 2000
     },
     {
       "epoch": 0.04,
-      "eval_accuracy": 0.2694227005870842,
-      "eval_loss": 4.3582682609558105,
-      "eval_runtime": 5.3878,
-      "eval_samples_per_second": 185.606,
-      "eval_steps_per_second": 1.485,
+      "eval_accuracy": 0.21657142857142858,
+      "eval_loss": 40.417686462402344,
+      "eval_runtime": 3.051,
+      "eval_samples_per_second": 40.97,
+      "eval_steps_per_second": 0.656,
       "step": 2000
     }
   ],
@@ -186,7 +186,7 @@
     }
   },
   "total_flos": 0.0,
-  "train_batch_size": 128,
+  "train_batch_size": 8,
   "trial_name": null,
   "trial_params": null
 }

last-checkpoint/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3ab34ea13b0e64af4c31a655731e444610b50d6a753b685ccf0d645cea5f76a3
-size 5265
+oid sha256:360066b4a5f1cc8c89941d389baaf9061071c4da2a624ff0fb9b3a9211c62ff7
+size 5905