| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.0005170965031348976, | |
| "eval_steps": 3, | |
| "global_step": 6, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 8.618275052248293e-05, | |
| "grad_norm": 1.301719069480896, | |
| "learning_rate": 2e-05, | |
| "loss": 1.7574, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 8.618275052248293e-05, | |
| "eval_loss": NaN, | |
| "eval_runtime": 197.1529, | |
| "eval_samples_per_second": 24.783, | |
| "eval_steps_per_second": 12.391, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.00017236550104496586, | |
| "grad_norm": 0.9337891340255737, | |
| "learning_rate": 4e-05, | |
| "loss": 1.6179, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.0002585482515674488, | |
| "grad_norm": 0.9951015114784241, | |
| "learning_rate": 6e-05, | |
| "loss": 1.3227, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0002585482515674488, | |
| "eval_loss": NaN, | |
| "eval_runtime": 197.0419, | |
| "eval_samples_per_second": 24.797, | |
| "eval_steps_per_second": 12.398, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0003447310020899317, | |
| "grad_norm": 1.4771511554718018, | |
| "learning_rate": 8e-05, | |
| "loss": 1.7685, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.0004309137526124146, | |
| "grad_norm": 1.1435108184814453, | |
| "learning_rate": 0.0001, | |
| "loss": 1.4952, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.0005170965031348976, | |
| "grad_norm": 1.2580000162124634, | |
| "learning_rate": 0.00012, | |
| "loss": 1.7866, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.0005170965031348976, | |
| "eval_loss": NaN, | |
| "eval_runtime": 197.1878, | |
| "eval_samples_per_second": 24.778, | |
| "eval_steps_per_second": 12.389, | |
| "step": 6 | |
| } | |
| ], | |
| "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": 676540148613120.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |