| { | |
| "best_global_step": null, | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 10.0, | |
| "eval_steps": 500, | |
| "global_step": 400, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.25, | |
| "grad_norm": 107.36187744140625, | |
| "learning_rate": 4.5e-06, | |
| "loss": 4.6121, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": 81.77188110351562, | |
| "learning_rate": 9.5e-06, | |
| "loss": 4.1314, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "grad_norm": 82.85482788085938, | |
| "learning_rate": 1.45e-05, | |
| "loss": 3.2993, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 63.60570526123047, | |
| "learning_rate": 1.95e-05, | |
| "loss": 2.3656, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "grad_norm": 46.48249816894531, | |
| "learning_rate": 1.95e-05, | |
| "loss": 1.2769, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "grad_norm": 44.416969299316406, | |
| "learning_rate": 1.8944444444444447e-05, | |
| "loss": 0.7619, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "grad_norm": 36.2529296875, | |
| "learning_rate": 1.838888888888889e-05, | |
| "loss": 0.4734, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 25.273788452148438, | |
| "learning_rate": 1.7833333333333334e-05, | |
| "loss": 0.3357, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 2.25, | |
| "grad_norm": 23.972837448120117, | |
| "learning_rate": 1.727777777777778e-05, | |
| "loss": 0.1776, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "grad_norm": 32.60515594482422, | |
| "learning_rate": 1.6722222222222225e-05, | |
| "loss": 0.173, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.75, | |
| "grad_norm": 11.92000961303711, | |
| "learning_rate": 1.616666666666667e-05, | |
| "loss": 0.1374, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 32.889869689941406, | |
| "learning_rate": 1.5611111111111113e-05, | |
| "loss": 0.1166, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 3.25, | |
| "grad_norm": 13.984763145446777, | |
| "learning_rate": 1.5055555555555556e-05, | |
| "loss": 0.1058, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 3.5, | |
| "grad_norm": 9.11774730682373, | |
| "learning_rate": 1.45e-05, | |
| "loss": 0.0893, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 3.75, | |
| "grad_norm": 10.404315948486328, | |
| "learning_rate": 1.3944444444444446e-05, | |
| "loss": 0.0994, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 39.227901458740234, | |
| "learning_rate": 1.338888888888889e-05, | |
| "loss": 0.0954, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 4.25, | |
| "grad_norm": 14.942120552062988, | |
| "learning_rate": 1.2833333333333335e-05, | |
| "loss": 0.0884, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 4.5, | |
| "grad_norm": 12.529537200927734, | |
| "learning_rate": 1.227777777777778e-05, | |
| "loss": 0.0724, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 4.75, | |
| "grad_norm": 7.07827091217041, | |
| "learning_rate": 1.1722222222222224e-05, | |
| "loss": 0.0805, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 26.651918411254883, | |
| "learning_rate": 1.1166666666666668e-05, | |
| "loss": 0.0919, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 5.25, | |
| "grad_norm": 13.577973365783691, | |
| "learning_rate": 1.0611111111111111e-05, | |
| "loss": 0.0732, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 5.5, | |
| "grad_norm": 12.209450721740723, | |
| "learning_rate": 1.0055555555555557e-05, | |
| "loss": 0.0648, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 5.75, | |
| "grad_norm": 33.70127487182617, | |
| "learning_rate": 9.5e-06, | |
| "loss": 0.0672, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 21.313919067382812, | |
| "learning_rate": 8.944444444444446e-06, | |
| "loss": 0.0742, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 6.25, | |
| "grad_norm": 21.87645721435547, | |
| "learning_rate": 8.38888888888889e-06, | |
| "loss": 0.0528, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 6.5, | |
| "grad_norm": 14.795717239379883, | |
| "learning_rate": 7.833333333333333e-06, | |
| "loss": 0.0646, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 6.75, | |
| "grad_norm": 17.657751083374023, | |
| "learning_rate": 7.277777777777778e-06, | |
| "loss": 0.0662, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 9.89352035522461, | |
| "learning_rate": 6.7222222222222235e-06, | |
| "loss": 0.0852, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 7.25, | |
| "grad_norm": 10.50017261505127, | |
| "learning_rate": 6.166666666666667e-06, | |
| "loss": 0.0672, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 7.5, | |
| "grad_norm": 16.115074157714844, | |
| "learning_rate": 5.611111111111112e-06, | |
| "loss": 0.0807, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 7.75, | |
| "grad_norm": 15.634113311767578, | |
| "learning_rate": 5.0555555555555555e-06, | |
| "loss": 0.056, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 4.4182329177856445, | |
| "learning_rate": 4.5e-06, | |
| "loss": 0.0576, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 8.25, | |
| "grad_norm": 22.3959903717041, | |
| "learning_rate": 3.944444444444445e-06, | |
| "loss": 0.0759, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 8.5, | |
| "grad_norm": 10.154661178588867, | |
| "learning_rate": 3.3888888888888893e-06, | |
| "loss": 0.0346, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 8.75, | |
| "grad_norm": 4.109118938446045, | |
| "learning_rate": 2.8333333333333335e-06, | |
| "loss": 0.0577, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 5.147540092468262, | |
| "learning_rate": 2.277777777777778e-06, | |
| "loss": 0.0587, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 9.25, | |
| "grad_norm": 8.975791931152344, | |
| "learning_rate": 1.7222222222222224e-06, | |
| "loss": 0.0507, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 9.5, | |
| "grad_norm": 25.286869049072266, | |
| "learning_rate": 1.1666666666666668e-06, | |
| "loss": 0.0496, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 9.75, | |
| "grad_norm": 13.948539733886719, | |
| "learning_rate": 6.111111111111112e-07, | |
| "loss": 0.0576, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 20.6357421875, | |
| "learning_rate": 5.555555555555556e-08, | |
| "loss": 0.08, | |
| "step": 400 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 400, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 2500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 0.0, | |
| "train_batch_size": 64, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |