| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.0003625231612019657, | |
| "eval_steps": 3, | |
| "global_step": 9, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 4.028035124466285e-05, | |
| "grad_norm": 2.2151663303375244, | |
| "learning_rate": 2e-05, | |
| "loss": 1.292, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 4.028035124466285e-05, | |
| "eval_loss": 1.337856650352478, | |
| "eval_runtime": 368.8154, | |
| "eval_samples_per_second": 28.345, | |
| "eval_steps_per_second": 14.172, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 8.05607024893257e-05, | |
| "grad_norm": 1.7885174751281738, | |
| "learning_rate": 4e-05, | |
| "loss": 1.1974, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.00012084105373398856, | |
| "grad_norm": 2.002810001373291, | |
| "learning_rate": 6e-05, | |
| "loss": 1.069, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.00012084105373398856, | |
| "eval_loss": 1.3132505416870117, | |
| "eval_runtime": 368.5438, | |
| "eval_samples_per_second": 28.366, | |
| "eval_steps_per_second": 14.183, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0001611214049786514, | |
| "grad_norm": 2.4847991466522217, | |
| "learning_rate": 8e-05, | |
| "loss": 1.365, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.00020140175622331428, | |
| "grad_norm": 2.2819066047668457, | |
| "learning_rate": 0.0001, | |
| "loss": 1.1394, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.00024168210746797713, | |
| "grad_norm": 2.7741761207580566, | |
| "learning_rate": 0.00012, | |
| "loss": 1.2401, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.00024168210746797713, | |
| "eval_loss": 1.0384124517440796, | |
| "eval_runtime": 368.0396, | |
| "eval_samples_per_second": 28.405, | |
| "eval_steps_per_second": 14.202, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.00028196245871264, | |
| "grad_norm": 2.818126678466797, | |
| "learning_rate": 0.00014, | |
| "loss": 1.0842, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.0003222428099573028, | |
| "grad_norm": 2.949310779571533, | |
| "learning_rate": 0.00016, | |
| "loss": 0.9708, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.0003625231612019657, | |
| "grad_norm": 2.6062393188476562, | |
| "learning_rate": 0.00018, | |
| "loss": 0.7602, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0003625231612019657, | |
| "eval_loss": 0.5603215098381042, | |
| "eval_runtime": 367.7961, | |
| "eval_samples_per_second": 28.423, | |
| "eval_steps_per_second": 14.212, | |
| "step": 9 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 10, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 3, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1719982274641920.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |