| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.0006819612506120165, | |
| "eval_steps": 10, | |
| "global_step": 39, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.748618591312863e-05, | |
| "eval_loss": 4.204365253448486, | |
| "eval_runtime": 3167.971, | |
| "eval_samples_per_second": 7.601, | |
| "eval_steps_per_second": 3.801, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 8.743092956564314e-05, | |
| "grad_norm": 5.006407737731934, | |
| "learning_rate": 5e-05, | |
| "loss": 4.6136, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.00017486185913128627, | |
| "grad_norm": 3.7020463943481445, | |
| "learning_rate": 0.0001, | |
| "loss": 3.1002, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.00017486185913128627, | |
| "eval_loss": 3.6543192863464355, | |
| "eval_runtime": 3212.9216, | |
| "eval_samples_per_second": 7.495, | |
| "eval_steps_per_second": 3.747, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.00026229278869692945, | |
| "grad_norm": 6.702174663543701, | |
| "learning_rate": 9.619397662556435e-05, | |
| "loss": 3.4858, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.00034972371826257255, | |
| "grad_norm": 8.694743156433105, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 2.7742, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.00034972371826257255, | |
| "eval_loss": 2.695681095123291, | |
| "eval_runtime": 3204.8183, | |
| "eval_samples_per_second": 7.514, | |
| "eval_steps_per_second": 3.757, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.0004371546478282157, | |
| "grad_norm": 5.783470153808594, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 2.3363, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.0005245855773938589, | |
| "grad_norm": 5.118088245391846, | |
| "learning_rate": 5e-05, | |
| "loss": 2.4692, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.0005245855773938589, | |
| "eval_loss": 2.5201780796051025, | |
| "eval_runtime": 3211.5638, | |
| "eval_samples_per_second": 7.498, | |
| "eval_steps_per_second": 3.749, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.000612016506959502, | |
| "grad_norm": 6.5606513023376465, | |
| "learning_rate": 3.086582838174551e-05, | |
| "loss": 2.5892, | |
| "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": 1.0265081577209856e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |