| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.16771488469601678, | |
| "eval_steps": 5, | |
| "global_step": 20, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.008385744234800839, | |
| "grad_norm": NaN, | |
| "learning_rate": 1e-05, | |
| "loss": 2176.5371, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.008385744234800839, | |
| "eval_loss": NaN, | |
| "eval_runtime": 170.8611, | |
| "eval_samples_per_second": 1.176, | |
| "eval_steps_per_second": 0.591, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.016771488469601678, | |
| "grad_norm": NaN, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.025157232704402517, | |
| "grad_norm": NaN, | |
| "learning_rate": 3e-05, | |
| "loss": 2.6692, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.033542976939203356, | |
| "grad_norm": NaN, | |
| "learning_rate": 4e-05, | |
| "loss": 0.0, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.041928721174004195, | |
| "grad_norm": NaN, | |
| "learning_rate": 5e-05, | |
| "loss": 0.0, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.041928721174004195, | |
| "eval_loss": NaN, | |
| "eval_runtime": 8.5114, | |
| "eval_samples_per_second": 23.615, | |
| "eval_steps_per_second": 11.866, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.050314465408805034, | |
| "grad_norm": NaN, | |
| "learning_rate": 6e-05, | |
| "loss": 0.0, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.05870020964360587, | |
| "grad_norm": NaN, | |
| "learning_rate": 7e-05, | |
| "loss": 2350.6379, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.06708595387840671, | |
| "grad_norm": NaN, | |
| "learning_rate": 8e-05, | |
| "loss": 0.0, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.07547169811320754, | |
| "grad_norm": NaN, | |
| "learning_rate": 9e-05, | |
| "loss": 0.0, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.08385744234800839, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.08385744234800839, | |
| "eval_loss": NaN, | |
| "eval_runtime": 8.5173, | |
| "eval_samples_per_second": 23.599, | |
| "eval_steps_per_second": 11.858, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.09224318658280922, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.755282581475769e-05, | |
| "loss": 0.2969, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.10062893081761007, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 0.1104, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.1090146750524109, | |
| "grad_norm": NaN, | |
| "learning_rate": 7.938926261462366e-05, | |
| "loss": 0.0, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.11740041928721175, | |
| "grad_norm": NaN, | |
| "learning_rate": 6.545084971874738e-05, | |
| "loss": 0.0, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.12578616352201258, | |
| "grad_norm": NaN, | |
| "learning_rate": 5e-05, | |
| "loss": 0.0, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.12578616352201258, | |
| "eval_loss": NaN, | |
| "eval_runtime": 8.5247, | |
| "eval_samples_per_second": 23.578, | |
| "eval_steps_per_second": 11.848, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.13417190775681342, | |
| "grad_norm": NaN, | |
| "learning_rate": 3.4549150281252636e-05, | |
| "loss": 0.0, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.14255765199161424, | |
| "grad_norm": NaN, | |
| "learning_rate": 2.061073738537635e-05, | |
| "loss": 0.0, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.1509433962264151, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.549150281252633e-06, | |
| "loss": 0.0, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.15932914046121593, | |
| "grad_norm": NaN, | |
| "learning_rate": 2.4471741852423237e-06, | |
| "loss": 67.9604, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.16771488469601678, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0, | |
| "loss": 0.0, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.16771488469601678, | |
| "eval_loss": NaN, | |
| "eval_runtime": 8.5325, | |
| "eval_samples_per_second": 23.557, | |
| "eval_steps_per_second": 11.837, | |
| "step": 20 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 20, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 5, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2.848328942144717e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |