| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.006347397973715262, | |
| "eval_steps": 10, | |
| "global_step": 39, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.00016275379419782723, | |
| "eval_loss": 3.2320632934570312, | |
| "eval_runtime": 30.9159, | |
| "eval_samples_per_second": 83.711, | |
| "eval_steps_per_second": 41.856, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0008137689709891362, | |
| "grad_norm": 234.95375061035156, | |
| "learning_rate": 5e-05, | |
| "loss": 12.9234, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.0016275379419782724, | |
| "grad_norm": 265.54571533203125, | |
| "learning_rate": 0.0001, | |
| "loss": 13.0763, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.0016275379419782724, | |
| "eval_loss": 3.215762138366699, | |
| "eval_runtime": 30.0768, | |
| "eval_samples_per_second": 86.046, | |
| "eval_steps_per_second": 43.023, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.0024413069129674086, | |
| "grad_norm": 481.0985412597656, | |
| "learning_rate": 9.619397662556435e-05, | |
| "loss": 13.2175, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.0032550758839565447, | |
| "grad_norm": 396.6003112792969, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 13.7691, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.0032550758839565447, | |
| "eval_loss": 3.1425418853759766, | |
| "eval_runtime": 30.1983, | |
| "eval_samples_per_second": 85.7, | |
| "eval_steps_per_second": 42.85, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.0040688448549456805, | |
| "grad_norm": 197.60797119140625, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 12.2875, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.004882613825934817, | |
| "grad_norm": 300.2901611328125, | |
| "learning_rate": 5e-05, | |
| "loss": 12.5609, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.004882613825934817, | |
| "eval_loss": 3.099902868270874, | |
| "eval_runtime": 30.4752, | |
| "eval_samples_per_second": 84.922, | |
| "eval_steps_per_second": 42.461, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.005696382796923953, | |
| "grad_norm": 253.1979217529297, | |
| "learning_rate": 3.086582838174551e-05, | |
| "loss": 13.2348, | |
| "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": 120449752104960.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |