| { | |
| "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": 108.46660614013672, | |
| "learning_rate": 4.5e-06, | |
| "loss": 6.8268, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": 87.85498046875, | |
| "learning_rate": 9.5e-06, | |
| "loss": 6.2692, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "grad_norm": 77.33071899414062, | |
| "learning_rate": 1.45e-05, | |
| "loss": 5.4279, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 84.33023071289062, | |
| "learning_rate": 1.95e-05, | |
| "loss": 4.4944, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "grad_norm": 73.86274719238281, | |
| "learning_rate": 1.95e-05, | |
| "loss": 3.2618, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "grad_norm": 59.88399124145508, | |
| "learning_rate": 1.8944444444444447e-05, | |
| "loss": 2.4952, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "grad_norm": 68.54973602294922, | |
| "learning_rate": 1.838888888888889e-05, | |
| "loss": 1.9448, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 64.97733306884766, | |
| "learning_rate": 1.7833333333333334e-05, | |
| "loss": 1.6445, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 2.25, | |
| "grad_norm": 58.61252212524414, | |
| "learning_rate": 1.727777777777778e-05, | |
| "loss": 1.1984, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "grad_norm": 39.56243133544922, | |
| "learning_rate": 1.6722222222222225e-05, | |
| "loss": 1.1865, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.75, | |
| "grad_norm": 48.185123443603516, | |
| "learning_rate": 1.616666666666667e-05, | |
| "loss": 1.0324, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 49.13726043701172, | |
| "learning_rate": 1.5611111111111113e-05, | |
| "loss": 0.8747, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 3.25, | |
| "grad_norm": 31.3487606048584, | |
| "learning_rate": 1.5055555555555556e-05, | |
| "loss": 0.7836, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 3.5, | |
| "grad_norm": 43.31613540649414, | |
| "learning_rate": 1.45e-05, | |
| "loss": 0.729, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 3.75, | |
| "grad_norm": 45.93892288208008, | |
| "learning_rate": 1.3944444444444446e-05, | |
| "loss": 0.7182, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 65.34429931640625, | |
| "learning_rate": 1.338888888888889e-05, | |
| "loss": 0.6853, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 4.25, | |
| "grad_norm": 40.68680191040039, | |
| "learning_rate": 1.2833333333333335e-05, | |
| "loss": 0.6082, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 4.5, | |
| "grad_norm": 23.430301666259766, | |
| "learning_rate": 1.227777777777778e-05, | |
| "loss": 0.6027, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 4.75, | |
| "grad_norm": 31.279895782470703, | |
| "learning_rate": 1.1722222222222224e-05, | |
| "loss": 0.5536, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 40.851078033447266, | |
| "learning_rate": 1.1166666666666668e-05, | |
| "loss": 0.5974, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 5.25, | |
| "grad_norm": 41.20182800292969, | |
| "learning_rate": 1.0611111111111111e-05, | |
| "loss": 0.4913, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 5.5, | |
| "grad_norm": 44.764102935791016, | |
| "learning_rate": 1.0055555555555557e-05, | |
| "loss": 0.5283, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 5.75, | |
| "grad_norm": 76.29002380371094, | |
| "learning_rate": 9.5e-06, | |
| "loss": 0.5435, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 42.7576904296875, | |
| "learning_rate": 8.944444444444446e-06, | |
| "loss": 0.5126, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 6.25, | |
| "grad_norm": 31.921890258789062, | |
| "learning_rate": 8.38888888888889e-06, | |
| "loss": 0.4732, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 6.5, | |
| "grad_norm": 344.25213623046875, | |
| "learning_rate": 7.833333333333333e-06, | |
| "loss": 0.507, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 6.75, | |
| "grad_norm": 39.58772659301758, | |
| "learning_rate": 7.277777777777778e-06, | |
| "loss": 0.4971, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 32.24217987060547, | |
| "learning_rate": 6.7222222222222235e-06, | |
| "loss": 0.4867, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 7.25, | |
| "grad_norm": 36.557579040527344, | |
| "learning_rate": 6.166666666666667e-06, | |
| "loss": 0.417, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 7.5, | |
| "grad_norm": 32.9907341003418, | |
| "learning_rate": 5.611111111111112e-06, | |
| "loss": 0.4371, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 7.75, | |
| "grad_norm": 48.47484588623047, | |
| "learning_rate": 5.0555555555555555e-06, | |
| "loss": 0.4747, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 39.65238952636719, | |
| "learning_rate": 4.5e-06, | |
| "loss": 0.4309, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 8.25, | |
| "grad_norm": 30.107986450195312, | |
| "learning_rate": 3.944444444444445e-06, | |
| "loss": 0.4543, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 8.5, | |
| "grad_norm": 34.437103271484375, | |
| "learning_rate": 3.3888888888888893e-06, | |
| "loss": 0.4244, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 8.75, | |
| "grad_norm": 43.94343566894531, | |
| "learning_rate": 2.8333333333333335e-06, | |
| "loss": 0.4722, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 36.12664031982422, | |
| "learning_rate": 2.277777777777778e-06, | |
| "loss": 0.438, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 9.25, | |
| "grad_norm": 43.92878341674805, | |
| "learning_rate": 1.7222222222222224e-06, | |
| "loss": 0.4211, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 9.5, | |
| "grad_norm": 35.80113983154297, | |
| "learning_rate": 1.1666666666666668e-06, | |
| "loss": 0.419, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 9.75, | |
| "grad_norm": 33.334144592285156, | |
| "learning_rate": 6.111111111111112e-07, | |
| "loss": 0.4148, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 33.29014205932617, | |
| "learning_rate": 5.555555555555556e-08, | |
| "loss": 0.4525, | |
| "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 | |
| } | |