| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.0, | |
| "eval_steps": 50, | |
| "global_step": 222, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.06756756756756757, | |
| "grad_norm": 18.147648336138747, | |
| "learning_rate": 5e-07, | |
| "loss": 1.7341, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.13513513513513514, | |
| "grad_norm": 12.567196378993486, | |
| "learning_rate": 1e-06, | |
| "loss": 1.5944, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.20270270270270271, | |
| "grad_norm": 7.429037368072137, | |
| "learning_rate": 9.98628141419305e-07, | |
| "loss": 1.2845, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.2702702702702703, | |
| "grad_norm": 4.009511516444553, | |
| "learning_rate": 9.94520093661082e-07, | |
| "loss": 1.1305, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.33783783783783783, | |
| "grad_norm": 4.23465040926726, | |
| "learning_rate": 9.876983993675989e-07, | |
| "loss": 1.0578, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.40540540540540543, | |
| "grad_norm": 3.5571250740361657, | |
| "learning_rate": 9.78200492138261e-07, | |
| "loss": 1.0215, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.47297297297297297, | |
| "grad_norm": 3.8692043296744756, | |
| "learning_rate": 9.66078491115194e-07, | |
| "loss": 0.9831, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.5405405405405406, | |
| "grad_norm": 3.6835209088708436, | |
| "learning_rate": 9.513989149828717e-07, | |
| "loss": 0.9719, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.6081081081081081, | |
| "grad_norm": 3.7600724565219084, | |
| "learning_rate": 9.342423169512071e-07, | |
| "loss": 0.9608, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.6756756756756757, | |
| "grad_norm": 3.666473757830759, | |
| "learning_rate": 9.147028427251009e-07, | |
| "loss": 0.934, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.6756756756756757, | |
| "eval_loss": 0.9346795082092285, | |
| "eval_runtime": 36.7724, | |
| "eval_samples_per_second": 57.108, | |
| "eval_steps_per_second": 0.897, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.7432432432432432, | |
| "grad_norm": 3.693572944803997, | |
| "learning_rate": 8.928877138860706e-07, | |
| "loss": 0.9282, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.8108108108108109, | |
| "grad_norm": 3.573512958166061, | |
| "learning_rate": 8.689166395208636e-07, | |
| "loss": 0.9244, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.8783783783783784, | |
| "grad_norm": 3.471592052796692, | |
| "learning_rate": 8.429211593257052e-07, | |
| "loss": 0.9272, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.9459459459459459, | |
| "grad_norm": 3.53208759809826, | |
| "learning_rate": 8.150439217908556e-07, | |
| "loss": 0.9078, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.0135135135135136, | |
| "grad_norm": 3.6886968178261696, | |
| "learning_rate": 7.854379014263876e-07, | |
| "loss": 0.8899, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 1.0810810810810811, | |
| "grad_norm": 3.541166696620505, | |
| "learning_rate": 7.542655593246103e-07, | |
| "loss": 0.8386, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.1486486486486487, | |
| "grad_norm": 3.581160863497663, | |
| "learning_rate": 7.216979516654943e-07, | |
| "loss": 0.8301, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 1.2162162162162162, | |
| "grad_norm": 3.6176679783753913, | |
| "learning_rate": 6.87913791057119e-07, | |
| "loss": 0.8229, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.2837837837837838, | |
| "grad_norm": 3.6731631175879484, | |
| "learning_rate": 6.530984658619733e-07, | |
| "loss": 0.8438, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 1.3513513513513513, | |
| "grad_norm": 3.6731091643467257, | |
| "learning_rate": 6.174430228904919e-07, | |
| "loss": 0.8288, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.3513513513513513, | |
| "eval_loss": 0.8865894079208374, | |
| "eval_runtime": 36.3701, | |
| "eval_samples_per_second": 57.74, | |
| "eval_steps_per_second": 0.907, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.4189189189189189, | |
| "grad_norm": 3.5335128896448253, | |
| "learning_rate": 5.8114311904423e-07, | |
| "loss": 0.8215, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 1.4864864864864864, | |
| "grad_norm": 3.690269611039034, | |
| "learning_rate": 5.443979476614674e-07, | |
| "loss": 0.8085, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 1.554054054054054, | |
| "grad_norm": 3.359019393396789, | |
| "learning_rate": 5.074091454568463e-07, | |
| "loss": 0.8189, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 1.6216216216216215, | |
| "grad_norm": 3.424018750837392, | |
| "learning_rate": 4.703796860531429e-07, | |
| "loss": 0.8138, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.689189189189189, | |
| "grad_norm": 3.3744478759084444, | |
| "learning_rate": 4.3351276617684285e-07, | |
| "loss": 0.8233, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 1.7567567567567568, | |
| "grad_norm": 3.816388446902284, | |
| "learning_rate": 3.970106906294509e-07, | |
| "loss": 0.8085, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.8243243243243243, | |
| "grad_norm": 3.5276501960621562, | |
| "learning_rate": 3.610737621531781e-07, | |
| "loss": 0.8133, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 1.8918918918918919, | |
| "grad_norm": 3.354366726555893, | |
| "learning_rate": 3.2589918228280066e-07, | |
| "loss": 0.7966, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.9594594594594594, | |
| "grad_norm": 3.5645103306323778, | |
| "learning_rate": 2.916799692151884e-07, | |
| "loss": 0.7986, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 2.027027027027027, | |
| "grad_norm": 3.787674005256534, | |
| "learning_rate": 2.5860389863462763e-07, | |
| "loss": 0.7923, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 2.027027027027027, | |
| "eval_loss": 0.8694743514060974, | |
| "eval_runtime": 36.3574, | |
| "eval_samples_per_second": 57.76, | |
| "eval_steps_per_second": 0.908, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 2.0945945945945947, | |
| "grad_norm": 3.552157993728247, | |
| "learning_rate": 2.2685247330608414e-07, | |
| "loss": 0.7511, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 2.1621621621621623, | |
| "grad_norm": 3.6209658799432995, | |
| "learning_rate": 1.9659992709070344e-07, | |
| "loss": 0.7588, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 2.22972972972973, | |
| "grad_norm": 3.6235800119871393, | |
| "learning_rate": 1.6801226884893893e-07, | |
| "loss": 0.7341, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 2.2972972972972974, | |
| "grad_norm": 3.839614745836778, | |
| "learning_rate": 1.412463714778343e-07, | |
| "loss": 0.7554, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 2.364864864864865, | |
| "grad_norm": 3.483048087846688, | |
| "learning_rate": 1.1644911108130434e-07, | |
| "loss": 0.7544, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 2.4324324324324325, | |
| "grad_norm": 3.6238622408287506, | |
| "learning_rate": 9.375656099715934e-08, | |
| "loss": 0.7555, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "grad_norm": 3.427837734720163, | |
| "learning_rate": 7.329324510360269e-08, | |
| "loss": 0.7524, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 2.5675675675675675, | |
| "grad_norm": 3.6888537439396067, | |
| "learning_rate": 5.517145450262639e-08, | |
| "loss": 0.7545, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 2.635135135135135, | |
| "grad_norm": 3.691825967842, | |
| "learning_rate": 3.9490631329964554e-08, | |
| "loss": 0.7583, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 2.7027027027027026, | |
| "grad_norm": 3.621833896748127, | |
| "learning_rate": 2.63368230729043e-08, | |
| "loss": 0.7615, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.7027027027027026, | |
| "eval_loss": 0.8699440956115723, | |
| "eval_runtime": 36.32, | |
| "eval_samples_per_second": 57.819, | |
| "eval_steps_per_second": 0.909, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.77027027027027, | |
| "grad_norm": 3.731982164876763, | |
| "learning_rate": 1.5782210390350713e-08, | |
| "loss": 0.7678, | |
| "step": 205 | |
| }, | |
| { | |
| "epoch": 2.8378378378378377, | |
| "grad_norm": 3.6719647864459666, | |
| "learning_rate": 7.884711026201584e-09, | |
| "loss": 0.7493, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 2.9054054054054053, | |
| "grad_norm": 3.5815863584732908, | |
| "learning_rate": 2.687661989531964e-09, | |
| "loss": 0.7561, | |
| "step": 215 | |
| }, | |
| { | |
| "epoch": 2.972972972972973, | |
| "grad_norm": 3.5948927539659783, | |
| "learning_rate": 2.1958174560282594e-10, | |
| "loss": 0.7664, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "step": 222, | |
| "total_flos": 1308778674782208.0, | |
| "train_loss": 0.8867547136169296, | |
| "train_runtime": 3082.1425, | |
| "train_samples_per_second": 18.396, | |
| "train_steps_per_second": 0.072 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 222, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 10000, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1308778674782208.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |