| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.0010177728585422949, | |
| "eval_steps": 5, | |
| "global_step": 20, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 5.088864292711474e-05, | |
| "grad_norm": NaN, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 5.088864292711474e-05, | |
| "eval_loss": NaN, | |
| "eval_runtime": 40.6823, | |
| "eval_samples_per_second": 203.405, | |
| "eval_steps_per_second": 101.715, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.00010177728585422949, | |
| "grad_norm": NaN, | |
| "learning_rate": 4e-05, | |
| "loss": 0.0, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.00015266592878134422, | |
| "grad_norm": NaN, | |
| "learning_rate": 6e-05, | |
| "loss": 0.0, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.00020355457170845897, | |
| "grad_norm": NaN, | |
| "learning_rate": 8e-05, | |
| "loss": 0.0, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.0002544432146355737, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.0002544432146355737, | |
| "eval_loss": NaN, | |
| "eval_runtime": 39.5272, | |
| "eval_samples_per_second": 209.35, | |
| "eval_steps_per_second": 104.687, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.00030533185756268844, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00012, | |
| "loss": 0.0, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.00035622050048980317, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00014, | |
| "loss": 0.0, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.00040710914341691794, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00016, | |
| "loss": 0.0, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.00045799778634403267, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00018, | |
| "loss": 0.0, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0005088864292711474, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.0005088864292711474, | |
| "eval_loss": NaN, | |
| "eval_runtime": 39.3434, | |
| "eval_samples_per_second": 210.328, | |
| "eval_steps_per_second": 105.177, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.0005597750721982621, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00019510565162951537, | |
| "loss": 0.0, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.0006106637151253769, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00018090169943749476, | |
| "loss": 0.0, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.0006615523580524917, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00015877852522924732, | |
| "loss": 0.0, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.0007124410009796063, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00013090169943749476, | |
| "loss": 0.0, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.0007633296439067211, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.0007633296439067211, | |
| "eval_loss": NaN, | |
| "eval_runtime": 39.492, | |
| "eval_samples_per_second": 209.536, | |
| "eval_steps_per_second": 104.781, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.0008142182868338359, | |
| "grad_norm": NaN, | |
| "learning_rate": 6.909830056250527e-05, | |
| "loss": 0.0, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.0008651069297609506, | |
| "grad_norm": NaN, | |
| "learning_rate": 4.12214747707527e-05, | |
| "loss": 0.0, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.0009159955726880653, | |
| "grad_norm": NaN, | |
| "learning_rate": 1.9098300562505266e-05, | |
| "loss": 0.0, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.0009668842156151801, | |
| "grad_norm": NaN, | |
| "learning_rate": 4.8943483704846475e-06, | |
| "loss": 0.0, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.0010177728585422949, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0, | |
| "loss": 0.0, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.0010177728585422949, | |
| "eval_loss": NaN, | |
| "eval_runtime": 39.886, | |
| "eval_samples_per_second": 207.466, | |
| "eval_steps_per_second": 103.746, | |
| "step": 20 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 20, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 5, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2066743296000.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |