| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.0002497589132712174, | |
| "eval_steps": 3, | |
| "global_step": 9, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 2.7750990363468595e-05, | |
| "grad_norm": 3.6328177452087402, | |
| "learning_rate": 2e-05, | |
| "loss": 3.0125, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 2.7750990363468595e-05, | |
| "eval_loss": 3.148585319519043, | |
| "eval_runtime": 652.9608, | |
| "eval_samples_per_second": 23.237, | |
| "eval_steps_per_second": 11.619, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 5.550198072693719e-05, | |
| "grad_norm": 4.217474937438965, | |
| "learning_rate": 4e-05, | |
| "loss": 2.7947, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 8.325297109040578e-05, | |
| "grad_norm": 3.2326135635375977, | |
| "learning_rate": 6e-05, | |
| "loss": 2.4977, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 8.325297109040578e-05, | |
| "eval_loss": 3.104830503463745, | |
| "eval_runtime": 555.0491, | |
| "eval_samples_per_second": 27.336, | |
| "eval_steps_per_second": 13.669, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.00011100396145387438, | |
| "grad_norm": 3.2247939109802246, | |
| "learning_rate": 8e-05, | |
| "loss": 2.6327, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.00013875495181734298, | |
| "grad_norm": 6.7599310874938965, | |
| "learning_rate": 0.0001, | |
| "loss": 4.3786, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.00016650594218081157, | |
| "grad_norm": 4.195891857147217, | |
| "learning_rate": 0.00012, | |
| "loss": 2.6846, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.00016650594218081157, | |
| "eval_loss": 2.576138973236084, | |
| "eval_runtime": 481.3293, | |
| "eval_samples_per_second": 31.523, | |
| "eval_steps_per_second": 15.763, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.00019425693254428018, | |
| "grad_norm": 4.960712909698486, | |
| "learning_rate": 0.00014, | |
| "loss": 2.8703, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.00022200792290774876, | |
| "grad_norm": 3.8892712593078613, | |
| "learning_rate": 0.00016, | |
| "loss": 2.3767, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.0002497589132712174, | |
| "grad_norm": 3.7098920345306396, | |
| "learning_rate": 0.00018, | |
| "loss": 1.813, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0002497589132712174, | |
| "eval_loss": 1.7475117444992065, | |
| "eval_runtime": 499.9676, | |
| "eval_samples_per_second": 30.348, | |
| "eval_steps_per_second": 15.175, | |
| "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": 450087597637632.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |