| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.32141422257934915, | |
| "eval_steps": 10, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0064282844515869825, | |
| "eval_loss": 2.085723400115967, | |
| "eval_runtime": 38.4863, | |
| "eval_samples_per_second": 27.23, | |
| "eval_steps_per_second": 3.404, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.06428284451586982, | |
| "grad_norm": 0.21534493565559387, | |
| "learning_rate": 9.330127018922194e-05, | |
| "loss": 2.0586, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.06428284451586982, | |
| "eval_loss": 2.0195319652557373, | |
| "eval_runtime": 38.3787, | |
| "eval_samples_per_second": 27.307, | |
| "eval_steps_per_second": 3.413, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.12856568903173965, | |
| "grad_norm": 0.17785143852233887, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 2.0029, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.12856568903173965, | |
| "eval_loss": 1.9852516651153564, | |
| "eval_runtime": 38.2636, | |
| "eval_samples_per_second": 27.389, | |
| "eval_steps_per_second": 3.424, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.19284853354760947, | |
| "grad_norm": 0.15681762993335724, | |
| "learning_rate": 3.705904774487396e-05, | |
| "loss": 1.9785, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.19284853354760947, | |
| "eval_loss": 1.9739582538604736, | |
| "eval_runtime": 38.2559, | |
| "eval_samples_per_second": 27.394, | |
| "eval_steps_per_second": 3.424, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.2571313780634793, | |
| "grad_norm": 0.17056262493133545, | |
| "learning_rate": 1.0332332985438248e-05, | |
| "loss": 1.982, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.2571313780634793, | |
| "eval_loss": 1.969922423362732, | |
| "eval_runtime": 38.2122, | |
| "eval_samples_per_second": 27.426, | |
| "eval_steps_per_second": 3.428, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.32141422257934915, | |
| "grad_norm": 0.1600278913974762, | |
| "learning_rate": 0.0, | |
| "loss": 1.9708, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.32141422257934915, | |
| "eval_loss": 1.9691742658615112, | |
| "eval_runtime": 38.2128, | |
| "eval_samples_per_second": 27.425, | |
| "eval_steps_per_second": 3.428, | |
| "step": 50 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "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": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 6.463267094396928e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |