| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 10.0, | |
| "eval_steps": 500, | |
| "global_step": 3180, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.9968553459119497, | |
| "grad_norm": 0.6745762825012207, | |
| "learning_rate": 2.7360787331269834e-05, | |
| "loss": 0.357, | |
| "step": 317 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.7412903225806452, | |
| "eval_loss": 0.14044521749019623, | |
| "eval_runtime": 5.4763, | |
| "eval_samples_per_second": 566.072, | |
| "eval_steps_per_second": 11.869, | |
| "step": 318 | |
| }, | |
| { | |
| "epoch": 1.9937106918238994, | |
| "grad_norm": 0.4204491376876831, | |
| "learning_rate": 2.433131838819874e-05, | |
| "loss": 0.1207, | |
| "step": 634 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8770967741935484, | |
| "eval_loss": 0.06287968158721924, | |
| "eval_runtime": 5.5283, | |
| "eval_samples_per_second": 560.754, | |
| "eval_steps_per_second": 11.758, | |
| "step": 636 | |
| }, | |
| { | |
| "epoch": 2.990566037735849, | |
| "grad_norm": 0.4077504873275757, | |
| "learning_rate": 2.130184944512765e-05, | |
| "loss": 0.0718, | |
| "step": 951 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.9116129032258065, | |
| "eval_loss": 0.04054585471749306, | |
| "eval_runtime": 5.4053, | |
| "eval_samples_per_second": 573.506, | |
| "eval_steps_per_second": 12.025, | |
| "step": 954 | |
| }, | |
| { | |
| "epoch": 3.9874213836477987, | |
| "grad_norm": 0.2433169037103653, | |
| "learning_rate": 1.8272380502056557e-05, | |
| "loss": 0.0534, | |
| "step": 1268 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.9274193548387096, | |
| "eval_loss": 0.03159501776099205, | |
| "eval_runtime": 5.4668, | |
| "eval_samples_per_second": 567.056, | |
| "eval_steps_per_second": 11.89, | |
| "step": 1272 | |
| }, | |
| { | |
| "epoch": 4.984276729559748, | |
| "grad_norm": 0.21281147003173828, | |
| "learning_rate": 1.5242911558985466e-05, | |
| "loss": 0.0442, | |
| "step": 1585 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.9309677419354838, | |
| "eval_loss": 0.02665688283741474, | |
| "eval_runtime": 5.5204, | |
| "eval_samples_per_second": 561.552, | |
| "eval_steps_per_second": 11.774, | |
| "step": 1590 | |
| }, | |
| { | |
| "epoch": 5.981132075471698, | |
| "grad_norm": 0.23512092232704163, | |
| "learning_rate": 1.2213442615914374e-05, | |
| "loss": 0.0389, | |
| "step": 1902 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.9329032258064516, | |
| "eval_loss": 0.024243181571364403, | |
| "eval_runtime": 5.4419, | |
| "eval_samples_per_second": 569.649, | |
| "eval_steps_per_second": 11.944, | |
| "step": 1908 | |
| }, | |
| { | |
| "epoch": 6.977987421383648, | |
| "grad_norm": 0.2756204605102539, | |
| "learning_rate": 9.18397367284328e-06, | |
| "loss": 0.0356, | |
| "step": 2219 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.9361290322580645, | |
| "eval_loss": 0.02270686812698841, | |
| "eval_runtime": 5.4205, | |
| "eval_samples_per_second": 571.9, | |
| "eval_steps_per_second": 11.991, | |
| "step": 2226 | |
| }, | |
| { | |
| "epoch": 7.9748427672955975, | |
| "grad_norm": 0.1520700752735138, | |
| "learning_rate": 6.154504729772188e-06, | |
| "loss": 0.0336, | |
| "step": 2536 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.9354838709677419, | |
| "eval_loss": 0.021294621750712395, | |
| "eval_runtime": 5.4685, | |
| "eval_samples_per_second": 566.883, | |
| "eval_steps_per_second": 11.886, | |
| "step": 2544 | |
| }, | |
| { | |
| "epoch": 8.971698113207546, | |
| "grad_norm": 0.2087375968694687, | |
| "learning_rate": 3.1250357867010953e-06, | |
| "loss": 0.0322, | |
| "step": 2853 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.9354838709677419, | |
| "eval_loss": 0.020687058568000793, | |
| "eval_runtime": 5.4663, | |
| "eval_samples_per_second": 567.112, | |
| "eval_steps_per_second": 11.891, | |
| "step": 2862 | |
| }, | |
| { | |
| "epoch": 9.968553459119496, | |
| "grad_norm": 0.14469225704669952, | |
| "learning_rate": 9.556684363000292e-08, | |
| "loss": 0.0314, | |
| "step": 3170 | |
| } | |
| ], | |
| "logging_steps": 317, | |
| "max_steps": 3180, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 1000000000.0, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 825254092458012.0, | |
| "train_batch_size": 48, | |
| "trial_name": null, | |
| "trial_params": { | |
| "alpha": 0.0478722237588074, | |
| "learning_rate": 3.0390256274340926e-05, | |
| "lr_scheduler_type": "linear", | |
| "num_train_epochs": 10, | |
| "temperature": 7.533742821161418, | |
| "weight_decay": 0.25236012891933407 | |
| } | |
| } | |