| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.0, | |
| "eval_steps": 500, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 177.15509899472193, | |
| "learning_rate": 2.0000000000000003e-06, | |
| "loss": 2.0908, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "grad_norm": 249.41431883318035, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 2.5319, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "grad_norm": 150.39879752385414, | |
| "learning_rate": 6e-06, | |
| "loss": 1.7308, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 37.728760587541345, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 0.9312, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 15.70356057623236, | |
| "learning_rate": 1e-05, | |
| "loss": 0.7705, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "grad_norm": 5.45224867064269, | |
| "learning_rate": 9.987820251299121e-06, | |
| "loss": 0.4923, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "grad_norm": 4.908282773077112, | |
| "learning_rate": 9.951340343707852e-06, | |
| "loss": 0.3728, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 3.502796679759215, | |
| "learning_rate": 9.890738003669029e-06, | |
| "loss": 0.2617, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "grad_norm": 1.7024726367446046, | |
| "learning_rate": 9.806308479691595e-06, | |
| "loss": 0.2508, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 0.9940466279699945, | |
| "learning_rate": 9.698463103929542e-06, | |
| "loss": 0.2137, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "grad_norm": 2.450501035506251, | |
| "learning_rate": 9.567727288213005e-06, | |
| "loss": 0.2401, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "grad_norm": 1.4427281883701062, | |
| "learning_rate": 9.414737964294636e-06, | |
| "loss": 0.2037, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "grad_norm": 0.904196965125475, | |
| "learning_rate": 9.24024048078213e-06, | |
| "loss": 0.1882, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "grad_norm": 0.9298794223496288, | |
| "learning_rate": 9.045084971874738e-06, | |
| "loss": 0.1844, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "grad_norm": 1.855575170826518, | |
| "learning_rate": 8.83022221559489e-06, | |
| "loss": 0.1994, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "grad_norm": 0.6710757494156826, | |
| "learning_rate": 8.596699001693257e-06, | |
| "loss": 0.1774, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "grad_norm": 2.329770545062615, | |
| "learning_rate": 8.345653031794292e-06, | |
| "loss": 0.1916, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "grad_norm": 0.928554494466913, | |
| "learning_rate": 8.078307376628292e-06, | |
| "loss": 0.1859, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "grad_norm": 0.9443531833751477, | |
| "learning_rate": 7.795964517353734e-06, | |
| "loss": 0.1699, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 0.7918115530539849, | |
| "learning_rate": 7.500000000000001e-06, | |
| "loss": 0.1783, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "grad_norm": 0.9196163123591217, | |
| "learning_rate": 7.191855733945388e-06, | |
| "loss": 0.167, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "grad_norm": 0.7595897977250279, | |
| "learning_rate": 6.873032967079562e-06, | |
| "loss": 0.1778, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "grad_norm": 0.6745291224887299, | |
| "learning_rate": 6.545084971874738e-06, | |
| "loss": 0.1546, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 0.8011676613761106, | |
| "learning_rate": 6.209609477998339e-06, | |
| "loss": 0.1674, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": 0.5793616119081132, | |
| "learning_rate": 5.8682408883346535e-06, | |
| "loss": 0.1489, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "grad_norm": 0.502456080126672, | |
| "learning_rate": 5.522642316338268e-06, | |
| "loss": 0.1673, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "grad_norm": 0.9029577834941641, | |
| "learning_rate": 5.174497483512506e-06, | |
| "loss": 0.1562, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "grad_norm": 0.5880447415239002, | |
| "learning_rate": 4.825502516487497e-06, | |
| "loss": 0.1538, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "grad_norm": 0.6959640251857308, | |
| "learning_rate": 4.477357683661734e-06, | |
| "loss": 0.1317, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "grad_norm": 0.819167997187879, | |
| "learning_rate": 4.131759111665349e-06, | |
| "loss": 0.159, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "grad_norm": 0.700213501190347, | |
| "learning_rate": 3.790390522001662e-06, | |
| "loss": 0.1332, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "grad_norm": 0.5453470463406256, | |
| "learning_rate": 3.4549150281252635e-06, | |
| "loss": 0.1493, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "grad_norm": 0.7267182413675126, | |
| "learning_rate": 3.12696703292044e-06, | |
| "loss": 0.1449, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "grad_norm": 0.6201985145724069, | |
| "learning_rate": 2.8081442660546126e-06, | |
| "loss": 0.1489, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "grad_norm": 0.6063378038267555, | |
| "learning_rate": 2.5000000000000015e-06, | |
| "loss": 0.1374, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "grad_norm": 0.6864441408653645, | |
| "learning_rate": 2.204035482646267e-06, | |
| "loss": 0.1326, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "grad_norm": 0.5642913934479935, | |
| "learning_rate": 1.9216926233717087e-06, | |
| "loss": 0.1117, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "grad_norm": 0.5716062286734344, | |
| "learning_rate": 1.6543469682057105e-06, | |
| "loss": 0.1294, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "grad_norm": 0.7133036398240772, | |
| "learning_rate": 1.4033009983067454e-06, | |
| "loss": 0.1449, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 0.5522465024093508, | |
| "learning_rate": 1.1697777844051105e-06, | |
| "loss": 0.1286, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "grad_norm": 0.5011972438671921, | |
| "learning_rate": 9.549150281252633e-07, | |
| "loss": 0.1347, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "grad_norm": 0.6161564938944896, | |
| "learning_rate": 7.597595192178702e-07, | |
| "loss": 0.1453, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "grad_norm": 0.6818018896853675, | |
| "learning_rate": 5.852620357053651e-07, | |
| "loss": 0.147, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "grad_norm": 0.6378587677381994, | |
| "learning_rate": 4.322727117869951e-07, | |
| "loss": 0.138, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "grad_norm": 0.47712509769630884, | |
| "learning_rate": 3.015368960704584e-07, | |
| "loss": 0.1273, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "grad_norm": 0.5347492598335275, | |
| "learning_rate": 1.9369152030840553e-07, | |
| "loss": 0.1372, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "grad_norm": 0.4147667548401041, | |
| "learning_rate": 1.0926199633097156e-07, | |
| "loss": 0.1365, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 0.45562021103374617, | |
| "learning_rate": 4.865965629214819e-08, | |
| "loss": 0.1365, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "grad_norm": 0.4580789267861981, | |
| "learning_rate": 1.2179748700879013e-08, | |
| "loss": 0.1165, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 0.4569340519935219, | |
| "learning_rate": 0.0, | |
| "loss": 0.1302, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.1237141489982605, | |
| "eval_runtime": 7.208, | |
| "eval_samples_per_second": 11.654, | |
| "eval_steps_per_second": 2.913, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "step": 50, | |
| "total_flos": 11109249515520.0, | |
| "train_loss": 0.31718048006296157, | |
| "train_runtime": 911.8037, | |
| "train_samples_per_second": 1.733, | |
| "train_steps_per_second": 0.055 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 50, | |
| "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": 11109249515520.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |