| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.00014824209581380793, | |
| "eval_steps": 3, | |
| "global_step": 9, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.6471343979311992e-05, | |
| "grad_norm": 49.8783073425293, | |
| "learning_rate": 2e-05, | |
| "loss": 6.0493, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 1.6471343979311992e-05, | |
| "eval_loss": 6.858318328857422, | |
| "eval_runtime": 652.2684, | |
| "eval_samples_per_second": 39.191, | |
| "eval_steps_per_second": 19.596, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 3.2942687958623985e-05, | |
| "grad_norm": 36.793212890625, | |
| "learning_rate": 4e-05, | |
| "loss": 6.6281, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 4.941403193793598e-05, | |
| "grad_norm": 30.170520782470703, | |
| "learning_rate": 6e-05, | |
| "loss": 7.7122, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 4.941403193793598e-05, | |
| "eval_loss": 6.623265266418457, | |
| "eval_runtime": 642.948, | |
| "eval_samples_per_second": 39.759, | |
| "eval_steps_per_second": 19.88, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 6.588537591724797e-05, | |
| "grad_norm": 35.831275939941406, | |
| "learning_rate": 8e-05, | |
| "loss": 6.2316, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 8.235671989655996e-05, | |
| "grad_norm": 17.99828338623047, | |
| "learning_rate": 0.0001, | |
| "loss": 5.8986, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 9.882806387587195e-05, | |
| "grad_norm": 13.067377090454102, | |
| "learning_rate": 0.00012, | |
| "loss": 4.9738, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 9.882806387587195e-05, | |
| "eval_loss": 5.341677188873291, | |
| "eval_runtime": 641.2785, | |
| "eval_samples_per_second": 39.863, | |
| "eval_steps_per_second": 19.932, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.00011529940785518395, | |
| "grad_norm": 20.99641227722168, | |
| "learning_rate": 0.00014, | |
| "loss": 5.6366, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.00013177075183449594, | |
| "grad_norm": 17.406648635864258, | |
| "learning_rate": 0.00016, | |
| "loss": 4.2952, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.00014824209581380793, | |
| "grad_norm": 16.842103958129883, | |
| "learning_rate": 0.00018, | |
| "loss": 4.0199, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.00014824209581380793, | |
| "eval_loss": 3.612816572189331, | |
| "eval_runtime": 649.2805, | |
| "eval_samples_per_second": 39.371, | |
| "eval_steps_per_second": 19.686, | |
| "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": 287765975531520.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |