| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.028901734104046242, | |
| "eval_steps": 25, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0005780346820809249, | |
| "eval_loss": 0.19159942865371704, | |
| "eval_runtime": 71.5012, | |
| "eval_samples_per_second": 2.699, | |
| "eval_steps_per_second": 2.699, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.002890173410404624, | |
| "grad_norm": 0.9248918294906616, | |
| "learning_rate": 5e-05, | |
| "loss": 0.2611, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.005780346820809248, | |
| "grad_norm": 0.45688870549201965, | |
| "learning_rate": 0.0001, | |
| "loss": 0.12, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.008670520231213872, | |
| "grad_norm": 0.4961397349834442, | |
| "learning_rate": 9.619397662556435e-05, | |
| "loss": 0.03, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.011560693641618497, | |
| "grad_norm": 1.3815051317214966, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 0.0308, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.014450867052023121, | |
| "grad_norm": 0.22113090753555298, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 0.0114, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.014450867052023121, | |
| "eval_loss": 0.022145938128232956, | |
| "eval_runtime": 71.3182, | |
| "eval_samples_per_second": 2.706, | |
| "eval_steps_per_second": 2.706, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.017341040462427744, | |
| "grad_norm": 0.04972805827856064, | |
| "learning_rate": 5e-05, | |
| "loss": 0.0181, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.02023121387283237, | |
| "grad_norm": 0.10225903242826462, | |
| "learning_rate": 3.086582838174551e-05, | |
| "loss": 0.0344, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.023121387283236993, | |
| "grad_norm": 0.15425756573677063, | |
| "learning_rate": 1.4644660940672627e-05, | |
| "loss": 0.0059, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.02601156069364162, | |
| "grad_norm": 0.3811052441596985, | |
| "learning_rate": 3.8060233744356633e-06, | |
| "loss": 0.0107, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.028901734104046242, | |
| "grad_norm": 0.34359514713287354, | |
| "learning_rate": 0.0, | |
| "loss": 0.0056, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.028901734104046242, | |
| "eval_loss": 0.016979020088911057, | |
| "eval_runtime": 71.5297, | |
| "eval_samples_per_second": 2.698, | |
| "eval_steps_per_second": 2.698, | |
| "step": 50 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 50, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 25, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 4636795522252800.0, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |