| { | |
| "best_global_step": null, | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 4.0, | |
| "eval_steps": 500, | |
| "global_step": 17560, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.11389521640091116, | |
| "grad_norm": 12.961915016174316, | |
| "learning_rate": 2.2733485193621873e-05, | |
| "loss": 2.138, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.22779043280182232, | |
| "grad_norm": 20.78866958618164, | |
| "learning_rate": 4.5512528473804106e-05, | |
| "loss": 0.8384, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.3416856492027335, | |
| "grad_norm": 15.086663246154785, | |
| "learning_rate": 6.829157175398633e-05, | |
| "loss": 0.6629, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.45558086560364464, | |
| "grad_norm": 34.28147888183594, | |
| "learning_rate": 7.99533422493134e-05, | |
| "loss": 0.6011, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.5694760820045558, | |
| "grad_norm": 42.04814529418945, | |
| "learning_rate": 7.95645044782997e-05, | |
| "loss": 0.5135, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.683371298405467, | |
| "grad_norm": 2.5099027156829834, | |
| "learning_rate": 7.878513717033947e-05, | |
| "loss": 0.49, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.7972665148063781, | |
| "grad_norm": 4.545470237731934, | |
| "learning_rate": 7.762293322991625e-05, | |
| "loss": 0.4222, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.9111617312072893, | |
| "grad_norm": 3.895155668258667, | |
| "learning_rate": 7.608936442882838e-05, | |
| "loss": 0.4221, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.0250569476082005, | |
| "grad_norm": 2.197298288345337, | |
| "learning_rate": 7.419956817171832e-05, | |
| "loss": 0.4035, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 1.1389521640091116, | |
| "grad_norm": 4.543397426605225, | |
| "learning_rate": 7.19721980792226e-05, | |
| "loss": 0.3188, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.2528473804100229, | |
| "grad_norm": 1.2027770280838013, | |
| "learning_rate": 6.942923986359271e-05, | |
| "loss": 0.301, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 1.366742596810934, | |
| "grad_norm": 5.7361955642700195, | |
| "learning_rate": 6.659579431422354e-05, | |
| "loss": 0.3052, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.4806378132118452, | |
| "grad_norm": 60.32481002807617, | |
| "learning_rate": 6.349982953517576e-05, | |
| "loss": 0.2945, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 1.5945330296127562, | |
| "grad_norm": 24.024532318115234, | |
| "learning_rate": 6.017190488028201e-05, | |
| "loss": 0.262, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 1.7084282460136673, | |
| "grad_norm": 5.258288860321045, | |
| "learning_rate": 5.664486931079191e-05, | |
| "loss": 0.2964, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 1.8223234624145785, | |
| "grad_norm": 22.747085571289062, | |
| "learning_rate": 5.295353715297871e-05, | |
| "loss": 0.2934, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 1.9362186788154898, | |
| "grad_norm": 1.9899779558181763, | |
| "learning_rate": 4.913434445620825e-05, | |
| "loss": 0.2535, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 2.050113895216401, | |
| "grad_norm": 0.44854220747947693, | |
| "learning_rate": 4.522498934345834e-05, | |
| "loss": 0.1939, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 2.164009111617312, | |
| "grad_norm": 12.502466201782227, | |
| "learning_rate": 4.126405990428233e-05, | |
| "loss": 0.161, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 2.277904328018223, | |
| "grad_norm": 0.9645262360572815, | |
| "learning_rate": 3.7290653303175396e-05, | |
| "loss": 0.1656, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 2.3917995444191344, | |
| "grad_norm": 0.041170768439769745, | |
| "learning_rate": 3.334398986301262e-05, | |
| "loss": 0.1394, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 2.5056947608200457, | |
| "grad_norm": 56.7460823059082, | |
| "learning_rate": 2.9463025932826876e-05, | |
| "loss": 0.1377, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 2.619589977220957, | |
| "grad_norm": 1.695715069770813, | |
| "learning_rate": 2.5686069361194757e-05, | |
| "loss": 0.1274, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 2.733485193621868, | |
| "grad_norm": 0.49590688943862915, | |
| "learning_rate": 2.205040137077874e-05, | |
| "loss": 0.1508, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 2.847380410022779, | |
| "grad_norm": 2.330212116241455, | |
| "learning_rate": 1.8591908566390646e-05, | |
| "loss": 0.13, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 2.9612756264236904, | |
| "grad_norm": 0.003335492219775915, | |
| "learning_rate": 1.5344728708915906e-05, | |
| "loss": 0.1254, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 3.075170842824601, | |
| "grad_norm": 0.9676417112350464, | |
| "learning_rate": 1.2340913751559621e-05, | |
| "loss": 0.0922, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 3.1890660592255125, | |
| "grad_norm": 0.6340539455413818, | |
| "learning_rate": 9.610113464484163e-06, | |
| "loss": 0.0876, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 3.3029612756264237, | |
| "grad_norm": 1.2587453126907349, | |
| "learning_rate": 7.179282770685981e-06, | |
| "loss": 0.0827, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 3.416856492027335, | |
| "grad_norm": 2.4137210845947266, | |
| "learning_rate": 5.072415681912745e-06, | |
| "loss": 0.0827, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 3.5307517084282463, | |
| "grad_norm": 1.391802191734314, | |
| "learning_rate": 3.3103084608609646e-06, | |
| "loss": 0.0803, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 3.644646924829157, | |
| "grad_norm": 0.5722363591194153, | |
| "learning_rate": 1.910354347409693e-06, | |
| "loss": 0.0777, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 3.7585421412300684, | |
| "grad_norm": 1.3810276985168457, | |
| "learning_rate": 8.863718750874395e-07, | |
| "loss": 0.0846, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 3.8724373576309796, | |
| "grad_norm": 0.8889294862747192, | |
| "learning_rate": 2.484684724094999e-07, | |
| "loss": 0.0867, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 3.9863325740318905, | |
| "grad_norm": 1.0857609510421753, | |
| "learning_rate": 2.940695436648433e-09, | |
| "loss": 0.0769, | |
| "step": 17500 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 17560, | |
| "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": 7.393629300129792e+16, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |