| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.853932584269663, | |
| "eval_steps": 500, | |
| "global_step": 88, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0898876404494382, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00019772727272727273, | |
| "loss": 9.4365, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.1797752808988764, | |
| "grad_norm": 10.061033248901367, | |
| "learning_rate": 0.0001931818181818182, | |
| "loss": 9.2158, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.2696629213483146, | |
| "grad_norm": 15.205401420593262, | |
| "learning_rate": 0.00018863636363636364, | |
| "loss": 8.8062, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.3595505617977528, | |
| "grad_norm": 20.087194442749023, | |
| "learning_rate": 0.00018409090909090909, | |
| "loss": 8.4059, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.449438202247191, | |
| "grad_norm": 12.80905818939209, | |
| "learning_rate": 0.00017954545454545456, | |
| "loss": 7.8943, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.5393258426966292, | |
| "grad_norm": 13.147987365722656, | |
| "learning_rate": 0.000175, | |
| "loss": 7.5915, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.6292134831460674, | |
| "grad_norm": 14.075699806213379, | |
| "learning_rate": 0.00017045454545454547, | |
| "loss": 7.3797, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.7191011235955056, | |
| "grad_norm": 6.000680923461914, | |
| "learning_rate": 0.00016590909090909094, | |
| "loss": 7.3359, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.8089887640449438, | |
| "grad_norm": 7.718422889709473, | |
| "learning_rate": 0.00016136363636363635, | |
| "loss": 7.3118, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.898876404494382, | |
| "grad_norm": 6.30356502532959, | |
| "learning_rate": 0.00015681818181818182, | |
| "loss": 7.2512, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.9887640449438202, | |
| "grad_norm": 11.91089153289795, | |
| "learning_rate": 0.00015227272727272727, | |
| "loss": 7.1725, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 1.0449438202247192, | |
| "grad_norm": 10.328221321105957, | |
| "learning_rate": 0.00014772727272727274, | |
| "loss": 4.4411, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 1.1348314606741572, | |
| "grad_norm": 11.233499526977539, | |
| "learning_rate": 0.0001431818181818182, | |
| "loss": 7.1546, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 1.2247191011235956, | |
| "grad_norm": 12.115806579589844, | |
| "learning_rate": 0.00013863636363636365, | |
| "loss": 7.146, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 1.3146067415730336, | |
| "grad_norm": 8.608111381530762, | |
| "learning_rate": 0.0001340909090909091, | |
| "loss": 7.1384, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.404494382022472, | |
| "grad_norm": 6.332569599151611, | |
| "learning_rate": 0.00012954545454545456, | |
| "loss": 7.0654, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 1.49438202247191, | |
| "grad_norm": 8.418745994567871, | |
| "learning_rate": 0.000125, | |
| "loss": 7.025, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 1.5842696629213484, | |
| "grad_norm": 3.712430000305176, | |
| "learning_rate": 0.00012045454545454546, | |
| "loss": 7.0801, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 1.6741573033707864, | |
| "grad_norm": 10.121654510498047, | |
| "learning_rate": 0.00011590909090909093, | |
| "loss": 7.1843, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 1.7640449438202248, | |
| "grad_norm": 4.086280345916748, | |
| "learning_rate": 0.00011136363636363636, | |
| "loss": 7.085, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.8539325842696628, | |
| "grad_norm": 6.2060933113098145, | |
| "learning_rate": 0.00010681818181818181, | |
| "loss": 7.0862, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 1.9438202247191012, | |
| "grad_norm": 3.9082467555999756, | |
| "learning_rate": 0.00010227272727272727, | |
| "loss": 7.034, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 3.6770272254943848, | |
| "learning_rate": 9.772727272727274e-05, | |
| "loss": 4.2733, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 2.0898876404494384, | |
| "grad_norm": 3.2819325923919678, | |
| "learning_rate": 9.318181818181818e-05, | |
| "loss": 7.0196, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 2.1797752808988764, | |
| "grad_norm": 6.796480655670166, | |
| "learning_rate": 8.863636363636364e-05, | |
| "loss": 7.0643, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 2.2696629213483144, | |
| "grad_norm": 5.090155124664307, | |
| "learning_rate": 8.40909090909091e-05, | |
| "loss": 6.9201, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 2.359550561797753, | |
| "grad_norm": 9.731218338012695, | |
| "learning_rate": 7.954545454545455e-05, | |
| "loss": 7.0715, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 2.449438202247191, | |
| "grad_norm": 5.852891445159912, | |
| "learning_rate": 7.500000000000001e-05, | |
| "loss": 7.1071, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 2.539325842696629, | |
| "grad_norm": 9.918388366699219, | |
| "learning_rate": 7.045454545454546e-05, | |
| "loss": 7.0041, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 2.629213483146067, | |
| "grad_norm": 5.5014190673828125, | |
| "learning_rate": 6.59090909090909e-05, | |
| "loss": 6.982, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 2.7191011235955056, | |
| "grad_norm": 7.316319465637207, | |
| "learning_rate": 6.136363636363636e-05, | |
| "loss": 6.9352, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 2.808988764044944, | |
| "grad_norm": 13.674769401550293, | |
| "learning_rate": 5.6818181818181825e-05, | |
| "loss": 6.9549, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 2.898876404494382, | |
| "grad_norm": 6.491578102111816, | |
| "learning_rate": 5.2272727272727274e-05, | |
| "loss": 7.0187, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 2.98876404494382, | |
| "grad_norm": 5.663166522979736, | |
| "learning_rate": 4.772727272727273e-05, | |
| "loss": 6.9697, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 3.044943820224719, | |
| "grad_norm": 5.13517427444458, | |
| "learning_rate": 4.318181818181819e-05, | |
| "loss": 4.3462, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 3.134831460674157, | |
| "grad_norm": 4.548388957977295, | |
| "learning_rate": 3.8636363636363636e-05, | |
| "loss": 6.8757, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 3.2247191011235956, | |
| "grad_norm": 7.8548431396484375, | |
| "learning_rate": 3.409090909090909e-05, | |
| "loss": 6.9019, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 3.3146067415730336, | |
| "grad_norm": 18.156105041503906, | |
| "learning_rate": 2.954545454545455e-05, | |
| "loss": 7.076, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 3.404494382022472, | |
| "grad_norm": 6.122278690338135, | |
| "learning_rate": 2.5e-05, | |
| "loss": 6.9759, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 3.49438202247191, | |
| "grad_norm": 8.052112579345703, | |
| "learning_rate": 2.0454545454545457e-05, | |
| "loss": 7.1131, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 3.5842696629213484, | |
| "grad_norm": 18.36809730529785, | |
| "learning_rate": 1.8181818181818182e-05, | |
| "loss": 6.8889, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 3.6741573033707864, | |
| "grad_norm": 3.2839643955230713, | |
| "learning_rate": 1.3636363636363637e-05, | |
| "loss": 6.9873, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 3.764044943820225, | |
| "grad_norm": 7.564036846160889, | |
| "learning_rate": 9.090909090909091e-06, | |
| "loss": 6.9841, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 3.853932584269663, | |
| "grad_norm": 3.23893404006958, | |
| "learning_rate": 4.5454545454545455e-06, | |
| "loss": 6.975, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 3.853932584269663, | |
| "step": 88, | |
| "total_flos": 343326692124408.0, | |
| "train_loss": 7.0837731144645, | |
| "train_runtime": 380.6612, | |
| "train_samples_per_second": 3.73, | |
| "train_steps_per_second": 0.231 | |
| } | |
| ], | |
| "logging_steps": 2, | |
| "max_steps": 88, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 4, | |
| "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": 343326692124408.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |