| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.001424783314204298, | |
| "eval_steps": 3, | |
| "global_step": 9, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0001583092571338109, | |
| "grad_norm": 1.5863964557647705, | |
| "learning_rate": 2e-05, | |
| "loss": 1.3986, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0001583092571338109, | |
| "eval_loss": 1.323094129562378, | |
| "eval_runtime": 79.7487, | |
| "eval_samples_per_second": 33.355, | |
| "eval_steps_per_second": 16.677, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0003166185142676218, | |
| "grad_norm": 1.0940614938735962, | |
| "learning_rate": 4e-05, | |
| "loss": 1.1118, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.0004749277714014327, | |
| "grad_norm": 1.3260576725006104, | |
| "learning_rate": 6e-05, | |
| "loss": 1.2353, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0004749277714014327, | |
| "eval_loss": 1.2929919958114624, | |
| "eval_runtime": 79.0791, | |
| "eval_samples_per_second": 33.637, | |
| "eval_steps_per_second": 16.819, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0006332370285352436, | |
| "grad_norm": 1.4070159196853638, | |
| "learning_rate": 8e-05, | |
| "loss": 1.3978, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.0007915462856690545, | |
| "grad_norm": 1.4026979207992554, | |
| "learning_rate": 0.0001, | |
| "loss": 1.1565, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.0009498555428028654, | |
| "grad_norm": 1.2851192951202393, | |
| "learning_rate": 0.00012, | |
| "loss": 1.0725, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.0009498555428028654, | |
| "eval_loss": 0.9489461779594421, | |
| "eval_runtime": 79.3249, | |
| "eval_samples_per_second": 33.533, | |
| "eval_steps_per_second": 16.766, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.0011081647999366764, | |
| "grad_norm": 1.2083780765533447, | |
| "learning_rate": 0.00014, | |
| "loss": 0.9325, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.0012664740570704873, | |
| "grad_norm": 1.3196492195129395, | |
| "learning_rate": 0.00016, | |
| "loss": 0.7352, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.001424783314204298, | |
| "grad_norm": 1.1015459299087524, | |
| "learning_rate": 0.00018, | |
| "loss": 0.7093, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.001424783314204298, | |
| "eval_loss": 0.5222042798995972, | |
| "eval_runtime": 79.494, | |
| "eval_samples_per_second": 33.462, | |
| "eval_steps_per_second": 16.731, | |
| "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": 1574605173030912.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |