| { | |
| "best_global_step": null, | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.8421052631578947, | |
| "eval_steps": 500, | |
| "global_step": 42, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.07017543859649122, | |
| "grad_norm": 10.52865219116211, | |
| "learning_rate": 0.0, | |
| "loss": 5.0782, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.14035087719298245, | |
| "grad_norm": 10.9165620803833, | |
| "learning_rate": 2.0000000000000003e-06, | |
| "loss": 5.1577, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.21052631578947367, | |
| "grad_norm": 11.1675443649292, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 5.1071, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.2807017543859649, | |
| "grad_norm": 10.448240280151367, | |
| "learning_rate": 6e-06, | |
| "loss": 5.1306, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.3508771929824561, | |
| "grad_norm": 10.703569412231445, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 5.0578, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.42105263157894735, | |
| "grad_norm": 10.621328353881836, | |
| "learning_rate": 1e-05, | |
| "loss": 4.9643, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.49122807017543857, | |
| "grad_norm": 10.412773132324219, | |
| "learning_rate": 1.2e-05, | |
| "loss": 4.8105, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.5614035087719298, | |
| "grad_norm": 10.143810272216797, | |
| "learning_rate": 1.4000000000000001e-05, | |
| "loss": 4.646, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.631578947368421, | |
| "grad_norm": 8.884021759033203, | |
| "learning_rate": 1.6000000000000003e-05, | |
| "loss": 4.478, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.7017543859649122, | |
| "grad_norm": 7.186990261077881, | |
| "learning_rate": 1.8e-05, | |
| "loss": 4.2099, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.7719298245614035, | |
| "grad_norm": 5.728164196014404, | |
| "learning_rate": 2e-05, | |
| "loss": 4.0316, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.8421052631578947, | |
| "grad_norm": 7.005011081695557, | |
| "learning_rate": 2.2000000000000003e-05, | |
| "loss": 3.9128, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.9122807017543859, | |
| "grad_norm": 8.047316551208496, | |
| "learning_rate": 2.4e-05, | |
| "loss": 3.7518, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.9824561403508771, | |
| "grad_norm": 6.397008895874023, | |
| "learning_rate": 2.6000000000000002e-05, | |
| "loss": 3.5923, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 4.87739372253418, | |
| "learning_rate": 2.8000000000000003e-05, | |
| "loss": 3.3723, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 1.0701754385964912, | |
| "grad_norm": 3.8167903423309326, | |
| "learning_rate": 3e-05, | |
| "loss": 3.3214, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 1.1403508771929824, | |
| "grad_norm": 3.506105899810791, | |
| "learning_rate": 3.2000000000000005e-05, | |
| "loss": 3.2114, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 1.2105263157894737, | |
| "grad_norm": 2.6847188472747803, | |
| "learning_rate": 3.4000000000000007e-05, | |
| "loss": 3.1914, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 1.280701754385965, | |
| "grad_norm": 2.0849263668060303, | |
| "learning_rate": 3.6e-05, | |
| "loss": 3.1213, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 1.3508771929824561, | |
| "grad_norm": 1.9912149906158447, | |
| "learning_rate": 3.8e-05, | |
| "loss": 3.0372, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.4210526315789473, | |
| "grad_norm": 1.7372136116027832, | |
| "learning_rate": 4e-05, | |
| "loss": 2.948, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 1.4912280701754386, | |
| "grad_norm": 1.7806267738342285, | |
| "learning_rate": 4.2e-05, | |
| "loss": 2.9269, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 1.5614035087719298, | |
| "grad_norm": 1.5322002172470093, | |
| "learning_rate": 4.4000000000000006e-05, | |
| "loss": 2.8877, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 1.631578947368421, | |
| "grad_norm": 1.6473444700241089, | |
| "learning_rate": 4.600000000000001e-05, | |
| "loss": 2.7834, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 1.7017543859649122, | |
| "grad_norm": 1.4832264184951782, | |
| "learning_rate": 4.8e-05, | |
| "loss": 2.7808, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 1.7719298245614035, | |
| "grad_norm": 1.4326348304748535, | |
| "learning_rate": 5e-05, | |
| "loss": 2.6874, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 1.8421052631578947, | |
| "grad_norm": 2.185389280319214, | |
| "learning_rate": 5.2000000000000004e-05, | |
| "loss": 2.667, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 1.912280701754386, | |
| "grad_norm": 4.019981384277344, | |
| "learning_rate": 5.4000000000000005e-05, | |
| "loss": 2.6309, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 1.9824561403508771, | |
| "grad_norm": 1.368363618850708, | |
| "learning_rate": 5.6000000000000006e-05, | |
| "loss": 2.5584, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 3.5658273696899414, | |
| "learning_rate": 5.8e-05, | |
| "loss": 2.4491, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.0701754385964914, | |
| "grad_norm": 2.8816487789154053, | |
| "learning_rate": 6e-05, | |
| "loss": 2.4853, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 2.1403508771929824, | |
| "grad_norm": 4.591163158416748, | |
| "learning_rate": 6.2e-05, | |
| "loss": 2.4324, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 2.2105263157894735, | |
| "grad_norm": 4.660356044769287, | |
| "learning_rate": 6.400000000000001e-05, | |
| "loss": 2.3511, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 2.280701754385965, | |
| "grad_norm": 2.6735446453094482, | |
| "learning_rate": 6.6e-05, | |
| "loss": 2.3241, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 2.3508771929824563, | |
| "grad_norm": 6.208858966827393, | |
| "learning_rate": 6.800000000000001e-05, | |
| "loss": 2.291, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 2.4210526315789473, | |
| "grad_norm": 3.5636632442474365, | |
| "learning_rate": 7e-05, | |
| "loss": 2.2498, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 2.4912280701754383, | |
| "grad_norm": 3.1954169273376465, | |
| "learning_rate": 7.2e-05, | |
| "loss": 2.23, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 2.56140350877193, | |
| "grad_norm": 3.8919479846954346, | |
| "learning_rate": 7.4e-05, | |
| "loss": 2.2094, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 2.6315789473684212, | |
| "grad_norm": 2.745952606201172, | |
| "learning_rate": 7.6e-05, | |
| "loss": 2.1835, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 2.7017543859649122, | |
| "grad_norm": 2.4738833904266357, | |
| "learning_rate": 7.800000000000001e-05, | |
| "loss": 2.1099, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 2.7719298245614032, | |
| "grad_norm": 5.345929145812988, | |
| "learning_rate": 8e-05, | |
| "loss": 2.0947, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 2.8421052631578947, | |
| "grad_norm": 2.14803147315979, | |
| "learning_rate": 8.2e-05, | |
| "loss": 2.0623, | |
| "step": 42 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 42, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 200, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 7.301010223949414e+17, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |