| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.7199999999999998, | |
| "eval_steps": 25, | |
| "global_step": 48, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 91.39633178710938, | |
| "learning_rate": 0.0002, | |
| "loss": 9.8814, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 69.82352447509766, | |
| "learning_rate": 0.0002, | |
| "loss": 9.508, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "grad_norm": 88.77470397949219, | |
| "learning_rate": 0.0002, | |
| "loss": 8.8757, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "grad_norm": 78.74607849121094, | |
| "learning_rate": 0.0002, | |
| "loss": 8.1277, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 57.297943115234375, | |
| "learning_rate": 0.0002, | |
| "loss": 7.5538, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 48.130455017089844, | |
| "learning_rate": 0.0002, | |
| "loss": 7.0051, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "grad_norm": 53.24027633666992, | |
| "learning_rate": 0.0002, | |
| "loss": 6.8888, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "grad_norm": 60.705360412597656, | |
| "learning_rate": 0.0002, | |
| "loss": 6.9736, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "grad_norm": 54.261192321777344, | |
| "learning_rate": 0.0002, | |
| "loss": 6.9427, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 26.81684684753418, | |
| "learning_rate": 0.0002, | |
| "loss": 6.5488, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "grad_norm": 26.71163558959961, | |
| "learning_rate": 0.0002, | |
| "loss": 6.1613, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 17.654388427734375, | |
| "learning_rate": 0.0002, | |
| "loss": 6.4909, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 11.796930313110352, | |
| "learning_rate": 0.0002, | |
| "loss": 3.1317, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "grad_norm": 12.123869895935059, | |
| "learning_rate": 0.0002, | |
| "loss": 6.5345, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "grad_norm": 34.23923873901367, | |
| "learning_rate": 0.0002, | |
| "loss": 5.8903, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "grad_norm": 22.84559440612793, | |
| "learning_rate": 0.0002, | |
| "loss": 5.7779, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "grad_norm": 13.740588188171387, | |
| "learning_rate": 0.0002, | |
| "loss": 6.2512, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "grad_norm": 12.967787742614746, | |
| "learning_rate": 0.0002, | |
| "loss": 5.9554, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 1.48, | |
| "grad_norm": 17.227495193481445, | |
| "learning_rate": 0.0002, | |
| "loss": 6.0653, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "grad_norm": 32.50784683227539, | |
| "learning_rate": 0.0002, | |
| "loss": 6.7432, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.6400000000000001, | |
| "grad_norm": 14.103690147399902, | |
| "learning_rate": 0.0002, | |
| "loss": 6.324, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 1.72, | |
| "grad_norm": 13.136197090148926, | |
| "learning_rate": 0.0002, | |
| "loss": 6.0088, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "grad_norm": 13.178682327270508, | |
| "learning_rate": 0.0002, | |
| "loss": 6.2658, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "grad_norm": 8.991060256958008, | |
| "learning_rate": 0.0002, | |
| "loss": 6.3406, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 1.96, | |
| "grad_norm": 11.259429931640625, | |
| "learning_rate": 0.0002, | |
| "loss": 6.4484, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 1.96, | |
| "eval_clap": 0.1398070603609085, | |
| "eval_loss": 5.790895462036133, | |
| "eval_runtime": 105.0758, | |
| "eval_samples_per_second": 0.152, | |
| "eval_steps_per_second": 0.152, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 5.0106096267700195, | |
| "learning_rate": 0.0002, | |
| "loss": 3.2002, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "grad_norm": 20.214805603027344, | |
| "learning_rate": 0.0002, | |
| "loss": 5.9964, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 2.16, | |
| "grad_norm": 9.884445190429688, | |
| "learning_rate": 0.0002, | |
| "loss": 6.158, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 2.24, | |
| "grad_norm": 29.28801918029785, | |
| "learning_rate": 0.0002, | |
| "loss": 5.6601, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 2.32, | |
| "grad_norm": 8.058592796325684, | |
| "learning_rate": 0.0002, | |
| "loss": 6.1302, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "grad_norm": 14.248044967651367, | |
| "learning_rate": 0.0002, | |
| "loss": 6.22, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 2.48, | |
| "grad_norm": 13.470457077026367, | |
| "learning_rate": 0.0002, | |
| "loss": 6.0524, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 2.56, | |
| "grad_norm": 14.079459190368652, | |
| "learning_rate": 0.0002, | |
| "loss": 6.2468, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 2.64, | |
| "grad_norm": 22.368366241455078, | |
| "learning_rate": 0.0002, | |
| "loss": 6.6833, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 2.7199999999999998, | |
| "grad_norm": 8.494449615478516, | |
| "learning_rate": 0.0002, | |
| "loss": 5.9105, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "grad_norm": 7.31801176071167, | |
| "learning_rate": 0.0002, | |
| "loss": 6.1115, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "grad_norm": 12.499704360961914, | |
| "learning_rate": 0.0002, | |
| "loss": 5.8303, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 2.96, | |
| "grad_norm": 12.041208267211914, | |
| "learning_rate": 0.0002, | |
| "loss": 6.2806, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 9.749319076538086, | |
| "learning_rate": 0.0002, | |
| "loss": 2.9795, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 3.08, | |
| "grad_norm": 7.609906196594238, | |
| "learning_rate": 0.0002, | |
| "loss": 6.2322, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 3.16, | |
| "grad_norm": 8.086981773376465, | |
| "learning_rate": 0.0002, | |
| "loss": 6.1848, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 3.24, | |
| "grad_norm": 22.7243709564209, | |
| "learning_rate": 0.0002, | |
| "loss": 5.6357, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 3.32, | |
| "grad_norm": 13.268460273742676, | |
| "learning_rate": 0.0002, | |
| "loss": 6.1597, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 3.4, | |
| "grad_norm": 9.500421524047852, | |
| "learning_rate": 0.0002, | |
| "loss": 6.0353, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 3.48, | |
| "grad_norm": 21.6617431640625, | |
| "learning_rate": 0.0002, | |
| "loss": 5.6887, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 3.56, | |
| "grad_norm": 14.880556106567383, | |
| "learning_rate": 0.0002, | |
| "loss": 5.6862, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 3.64, | |
| "grad_norm": 14.079529762268066, | |
| "learning_rate": 0.0002, | |
| "loss": 6.1141, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 3.7199999999999998, | |
| "grad_norm": 13.863551139831543, | |
| "learning_rate": 0.0002, | |
| "loss": 5.5132, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 3.7199999999999998, | |
| "step": 48, | |
| "total_flos": 84973802219964.0, | |
| "train_loss": 6.279264723261197, | |
| "train_runtime": 934.4841, | |
| "train_samples_per_second": 0.856, | |
| "train_steps_per_second": 0.051 | |
| } | |
| ], | |
| "logging_steps": 1.0, | |
| "max_steps": 48, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 4, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 84973802219964.0, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |