| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.000637603917176886, | |
| "eval_steps": 10, | |
| "global_step": 39, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.6348818389150925e-05, | |
| "eval_loss": 9.864830017089844, | |
| "eval_runtime": 1629.5514, | |
| "eval_samples_per_second": 15.805, | |
| "eval_steps_per_second": 7.903, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 8.174409194575462e-05, | |
| "grad_norm": 22.458637237548828, | |
| "learning_rate": 5e-05, | |
| "loss": 9.4677, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.00016348818389150924, | |
| "grad_norm": 18.819719314575195, | |
| "learning_rate": 0.0001, | |
| "loss": 7.4993, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.00016348818389150924, | |
| "eval_loss": 4.158893585205078, | |
| "eval_runtime": 1632.8948, | |
| "eval_samples_per_second": 15.773, | |
| "eval_steps_per_second": 7.887, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.00024523227583726385, | |
| "grad_norm": 17.953746795654297, | |
| "learning_rate": 9.619397662556435e-05, | |
| "loss": 2.3714, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.0003269763677830185, | |
| "grad_norm": 11.434354782104492, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 0.4002, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.0003269763677830185, | |
| "eval_loss": 0.4900224506855011, | |
| "eval_runtime": 1636.1371, | |
| "eval_samples_per_second": 15.741, | |
| "eval_steps_per_second": 7.871, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.0004087204597287731, | |
| "grad_norm": 13.020814895629883, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 0.3843, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.0004904645516745277, | |
| "grad_norm": 13.445178985595703, | |
| "learning_rate": 5e-05, | |
| "loss": 0.3962, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.0004904645516745277, | |
| "eval_loss": 0.3042037785053253, | |
| "eval_runtime": 1633.8448, | |
| "eval_samples_per_second": 15.763, | |
| "eval_steps_per_second": 7.882, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.0005722086436202823, | |
| "grad_norm": 9.87356948852539, | |
| "learning_rate": 3.086582838174551e-05, | |
| "loss": 0.2852, | |
| "step": 35 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 50, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 13, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 6835416091066368.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |