| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 4.0, | |
| "eval_steps": 500, | |
| "global_step": 750, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 5e-05, | |
| "loss": 1.3988, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.6660784313725491, | |
| "eval_loss": 1.2107256650924683, | |
| "eval_runtime": 7.728, | |
| "eval_samples_per_second": 64.7, | |
| "eval_steps_per_second": 8.152, | |
| "step": 187 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_exact_match": 19.4, | |
| "eval_f1": 27.911959489312448, | |
| "eval_qa_bleu": 2.842200543476061, | |
| "eval_qa_exact_match": 0.152, | |
| "eval_recite_bleu": 12.149078895446083, | |
| "eval_recite_exact_match": 0.0, | |
| "step": 187 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "learning_rate": 5e-05, | |
| "loss": 1.238, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 5e-05, | |
| "loss": 1.1977, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.6676078431372549, | |
| "eval_loss": 1.1987448930740356, | |
| "eval_runtime": 7.4628, | |
| "eval_samples_per_second": 66.999, | |
| "eval_steps_per_second": 8.442, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_exact_match": 21.4, | |
| "eval_f1": 27.979969946667268, | |
| "eval_qa_bleu": 3.4273838915653356, | |
| "eval_qa_exact_match": 0.158, | |
| "eval_recite_bleu": 11.474877690710539, | |
| "eval_recite_exact_match": 0.0, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 2.13, | |
| "learning_rate": 5e-05, | |
| "loss": 1.1628, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.67, | |
| "learning_rate": 5e-05, | |
| "loss": 1.1458, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.6675764705882353, | |
| "eval_loss": 1.1964411735534668, | |
| "eval_runtime": 6.6677, | |
| "eval_samples_per_second": 74.988, | |
| "eval_steps_per_second": 9.448, | |
| "step": 562 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_exact_match": 23.4, | |
| "eval_f1": 31.525761083856903, | |
| "eval_qa_bleu": 2.0606924929592334, | |
| "eval_qa_exact_match": 0.17, | |
| "eval_recite_bleu": 12.027903845545522, | |
| "eval_recite_exact_match": 0.0, | |
| "step": 562 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "learning_rate": 5e-05, | |
| "loss": 1.1179, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 3.73, | |
| "learning_rate": 5e-05, | |
| "loss": 1.0707, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.6665960784313726, | |
| "eval_loss": 1.208784580230713, | |
| "eval_runtime": 6.6546, | |
| "eval_samples_per_second": 75.136, | |
| "eval_steps_per_second": 9.467, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_exact_match": 23.4, | |
| "eval_f1": 33.18893657873498, | |
| "eval_qa_bleu": 3.8183981024520506, | |
| "eval_qa_exact_match": 0.176, | |
| "eval_recite_bleu": 14.071857337313705, | |
| "eval_recite_exact_match": 0.0, | |
| "step": 750 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 9350, | |
| "num_train_epochs": 50, | |
| "save_steps": 500, | |
| "total_flos": 1.738045862539428e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |