| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.9696969696969697, | |
| "eval_steps": 6, | |
| "global_step": 24, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.04040404040404041, | |
| "grad_norm": 0.1962890625, | |
| "learning_rate": 2e-05, | |
| "loss": 1.5483, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.04040404040404041, | |
| "eval_loss": 1.565699815750122, | |
| "eval_runtime": 10.0556, | |
| "eval_samples_per_second": 9.945, | |
| "eval_steps_per_second": 9.945, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.08080808080808081, | |
| "grad_norm": 0.1943359375, | |
| "learning_rate": 4e-05, | |
| "loss": 1.4502, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.12121212121212122, | |
| "grad_norm": 0.1962890625, | |
| "learning_rate": 6e-05, | |
| "loss": 1.416, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.16161616161616163, | |
| "grad_norm": 0.24609375, | |
| "learning_rate": 8e-05, | |
| "loss": 1.5064, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.20202020202020202, | |
| "grad_norm": 0.1982421875, | |
| "learning_rate": 0.0001, | |
| "loss": 1.448, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.24242424242424243, | |
| "grad_norm": 0.1767578125, | |
| "learning_rate": 0.00012, | |
| "loss": 1.3687, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.24242424242424243, | |
| "eval_loss": 1.4984517097473145, | |
| "eval_runtime": 10.4286, | |
| "eval_samples_per_second": 9.589, | |
| "eval_steps_per_second": 9.589, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.2828282828282828, | |
| "grad_norm": 0.1796875, | |
| "learning_rate": 0.00014, | |
| "loss": 1.2437, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.32323232323232326, | |
| "grad_norm": 0.146484375, | |
| "learning_rate": 0.00016, | |
| "loss": 1.2156, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.36363636363636365, | |
| "grad_norm": 0.1376953125, | |
| "learning_rate": 0.00018, | |
| "loss": 1.281, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.40404040404040403, | |
| "grad_norm": 0.1259765625, | |
| "learning_rate": 0.0002, | |
| "loss": 1.3515, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.4444444444444444, | |
| "grad_norm": 0.1337890625, | |
| "learning_rate": 0.00019749279121818235, | |
| "loss": 1.3942, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.48484848484848486, | |
| "grad_norm": 0.162109375, | |
| "learning_rate": 0.0001900968867902419, | |
| "loss": 1.1948, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.48484848484848486, | |
| "eval_loss": 1.2932639122009277, | |
| "eval_runtime": 10.761, | |
| "eval_samples_per_second": 9.293, | |
| "eval_steps_per_second": 9.293, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.5252525252525253, | |
| "grad_norm": 0.171875, | |
| "learning_rate": 0.000178183148246803, | |
| "loss": 1.2631, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.5656565656565656, | |
| "grad_norm": 0.1435546875, | |
| "learning_rate": 0.00016234898018587337, | |
| "loss": 1.2711, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.6060606060606061, | |
| "grad_norm": 0.12353515625, | |
| "learning_rate": 0.00014338837391175582, | |
| "loss": 1.2824, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.6464646464646465, | |
| "grad_norm": 0.115234375, | |
| "learning_rate": 0.00012225209339563145, | |
| "loss": 1.2682, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.6868686868686869, | |
| "grad_norm": 0.10009765625, | |
| "learning_rate": 0.0001, | |
| "loss": 1.1965, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.7272727272727273, | |
| "grad_norm": 0.11328125, | |
| "learning_rate": 7.774790660436858e-05, | |
| "loss": 1.2227, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.7272727272727273, | |
| "eval_loss": 1.2484949827194214, | |
| "eval_runtime": 10.7358, | |
| "eval_samples_per_second": 9.315, | |
| "eval_steps_per_second": 9.315, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.7676767676767676, | |
| "grad_norm": 0.11474609375, | |
| "learning_rate": 5.6611626088244194e-05, | |
| "loss": 1.1904, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.8080808080808081, | |
| "grad_norm": 0.0986328125, | |
| "learning_rate": 3.7651019814126654e-05, | |
| "loss": 1.2903, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.8484848484848485, | |
| "grad_norm": 0.1005859375, | |
| "learning_rate": 2.181685175319702e-05, | |
| "loss": 1.3272, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.8888888888888888, | |
| "grad_norm": 0.140625, | |
| "learning_rate": 9.903113209758096e-06, | |
| "loss": 1.2255, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.9292929292929293, | |
| "grad_norm": 0.0986328125, | |
| "learning_rate": 2.5072087818176382e-06, | |
| "loss": 1.2386, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.9696969696969697, | |
| "grad_norm": 0.107421875, | |
| "learning_rate": 0.0, | |
| "loss": 1.2483, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.9696969696969697, | |
| "eval_loss": 1.235521674156189, | |
| "eval_runtime": 10.7486, | |
| "eval_samples_per_second": 9.304, | |
| "eval_steps_per_second": 9.304, | |
| "step": 24 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 24, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2500249176244224.0, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |