| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.0006472142818618197, | |
| "eval_steps": 10, | |
| "global_step": 39, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.6595237996456916e-05, | |
| "eval_loss": 0.9319517016410828, | |
| "eval_runtime": 684.835, | |
| "eval_samples_per_second": 37.048, | |
| "eval_steps_per_second": 18.524, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 8.297618998228458e-05, | |
| "grad_norm": 3.183830499649048, | |
| "learning_rate": 5e-05, | |
| "loss": 3.624, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.00016595237996456917, | |
| "grad_norm": 4.00230598449707, | |
| "learning_rate": 0.0001, | |
| "loss": 3.7706, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.00016595237996456917, | |
| "eval_loss": 0.7170787453651428, | |
| "eval_runtime": 692.7042, | |
| "eval_samples_per_second": 36.627, | |
| "eval_steps_per_second": 18.314, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.00024892856994685375, | |
| "grad_norm": 2.8913145065307617, | |
| "learning_rate": 9.619397662556435e-05, | |
| "loss": 2.088, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.00033190475992913833, | |
| "grad_norm": 5.755887985229492, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 1.4491, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.00033190475992913833, | |
| "eval_loss": 0.2670595943927765, | |
| "eval_runtime": 689.8036, | |
| "eval_samples_per_second": 36.781, | |
| "eval_steps_per_second": 18.391, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.0004148809499114229, | |
| "grad_norm": 5.549856185913086, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 0.9362, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.0004978571398937075, | |
| "grad_norm": 7.281345844268799, | |
| "learning_rate": 5e-05, | |
| "loss": 0.5976, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.0004978571398937075, | |
| "eval_loss": 0.12552273273468018, | |
| "eval_runtime": 689.2345, | |
| "eval_samples_per_second": 36.812, | |
| "eval_steps_per_second": 18.406, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.0005808333298759921, | |
| "grad_norm": 5.342987060546875, | |
| "learning_rate": 3.086582838174551e-05, | |
| "loss": 0.488, | |
| "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": 1179360857948160.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |