| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.17805475183618963, | |
| "eval_steps": 13, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0035610950367237926, | |
| "grad_norm": 4.179213523864746, | |
| "learning_rate": 1e-05, | |
| "loss": 4.8864, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0035610950367237926, | |
| "eval_loss": 5.30247163772583, | |
| "eval_runtime": 126.7922, | |
| "eval_samples_per_second": 3.731, | |
| "eval_steps_per_second": 1.869, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.007122190073447585, | |
| "grad_norm": 3.5513782501220703, | |
| "learning_rate": 2e-05, | |
| "loss": 5.1702, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.010683285110171378, | |
| "grad_norm": 4.266094207763672, | |
| "learning_rate": 3e-05, | |
| "loss": 5.3837, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.01424438014689517, | |
| "grad_norm": 4.963681697845459, | |
| "learning_rate": 4e-05, | |
| "loss": 4.7062, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.01780547518361896, | |
| "grad_norm": 5.114255905151367, | |
| "learning_rate": 5e-05, | |
| "loss": 5.045, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.021366570220342756, | |
| "grad_norm": 6.496891498565674, | |
| "learning_rate": 6e-05, | |
| "loss": 5.0374, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.024927665257066547, | |
| "grad_norm": 8.91849422454834, | |
| "learning_rate": 7e-05, | |
| "loss": 4.6738, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.02848876029379034, | |
| "grad_norm": 4.901350498199463, | |
| "learning_rate": 8e-05, | |
| "loss": 4.7986, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.03204985533051413, | |
| "grad_norm": 3.618027925491333, | |
| "learning_rate": 9e-05, | |
| "loss": 3.4925, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.03561095036723792, | |
| "grad_norm": 3.863649368286133, | |
| "learning_rate": 0.0001, | |
| "loss": 4.3912, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.03917204540396172, | |
| "grad_norm": 4.39957857131958, | |
| "learning_rate": 9.98458666866564e-05, | |
| "loss": 3.5932, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.04273314044068551, | |
| "grad_norm": 5.253383636474609, | |
| "learning_rate": 9.938441702975689e-05, | |
| "loss": 3.0919, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.0462942354774093, | |
| "grad_norm": 2.535303831100464, | |
| "learning_rate": 9.861849601988383e-05, | |
| "loss": 2.5898, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.0462942354774093, | |
| "eval_loss": 2.55889630317688, | |
| "eval_runtime": 110.9541, | |
| "eval_samples_per_second": 4.263, | |
| "eval_steps_per_second": 2.136, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.049855330514133094, | |
| "grad_norm": 2.924900770187378, | |
| "learning_rate": 9.755282581475769e-05, | |
| "loss": 2.3642, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.05341642555085689, | |
| "grad_norm": 1.9190605878829956, | |
| "learning_rate": 9.619397662556435e-05, | |
| "loss": 2.245, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.05697752058758068, | |
| "grad_norm": 4.164839744567871, | |
| "learning_rate": 9.45503262094184e-05, | |
| "loss": 2.4275, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.06053861562430447, | |
| "grad_norm": 3.0086185932159424, | |
| "learning_rate": 9.263200821770461e-05, | |
| "loss": 2.5735, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.06409971066102826, | |
| "grad_norm": 2.273061513900757, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 2.2379, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.06766080569775205, | |
| "grad_norm": 1.9004080295562744, | |
| "learning_rate": 8.802029828000156e-05, | |
| "loss": 2.3375, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.07122190073447585, | |
| "grad_norm": 2.002903461456299, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 2.2203, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.07478299577119965, | |
| "grad_norm": 1.6309473514556885, | |
| "learning_rate": 8.247240241650918e-05, | |
| "loss": 2.1511, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.07834409080792344, | |
| "grad_norm": 1.8154593706130981, | |
| "learning_rate": 7.938926261462366e-05, | |
| "loss": 2.0112, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.08190518584464723, | |
| "grad_norm": 1.8795080184936523, | |
| "learning_rate": 7.612492823579745e-05, | |
| "loss": 1.8034, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.08546628088137102, | |
| "grad_norm": 1.9428141117095947, | |
| "learning_rate": 7.269952498697734e-05, | |
| "loss": 1.9395, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.08902737591809481, | |
| "grad_norm": 1.6623902320861816, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 2.1667, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.0925884709548186, | |
| "grad_norm": 1.343985915184021, | |
| "learning_rate": 6.545084971874738e-05, | |
| "loss": 2.2539, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.0925884709548186, | |
| "eval_loss": 2.0758728981018066, | |
| "eval_runtime": 110.911, | |
| "eval_samples_per_second": 4.265, | |
| "eval_steps_per_second": 2.137, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.0961495659915424, | |
| "grad_norm": 1.4864414930343628, | |
| "learning_rate": 6.167226819279528e-05, | |
| "loss": 2.1355, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.09971066102826619, | |
| "grad_norm": 1.316426396369934, | |
| "learning_rate": 5.782172325201155e-05, | |
| "loss": 1.9895, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.10327175606498998, | |
| "grad_norm": 1.437731146812439, | |
| "learning_rate": 5.392295478639225e-05, | |
| "loss": 2.0277, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.10683285110171378, | |
| "grad_norm": 1.660027027130127, | |
| "learning_rate": 5e-05, | |
| "loss": 2.228, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.11039394613843757, | |
| "grad_norm": 1.3694123029708862, | |
| "learning_rate": 4.607704521360776e-05, | |
| "loss": 1.9222, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.11395504117516136, | |
| "grad_norm": 1.639948844909668, | |
| "learning_rate": 4.2178276747988446e-05, | |
| "loss": 2.158, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.11751613621188516, | |
| "grad_norm": 2.648879051208496, | |
| "learning_rate": 3.832773180720475e-05, | |
| "loss": 2.0388, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.12107723124860895, | |
| "grad_norm": 1.5599783658981323, | |
| "learning_rate": 3.4549150281252636e-05, | |
| "loss": 2.4029, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.12463832628533274, | |
| "grad_norm": 1.3389352560043335, | |
| "learning_rate": 3.086582838174551e-05, | |
| "loss": 1.7365, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.12819942132205653, | |
| "grad_norm": 1.6567745208740234, | |
| "learning_rate": 2.7300475013022663e-05, | |
| "loss": 2.1814, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.13176051635878033, | |
| "grad_norm": 1.775384783744812, | |
| "learning_rate": 2.3875071764202563e-05, | |
| "loss": 2.3459, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.1353216113955041, | |
| "grad_norm": 1.3190244436264038, | |
| "learning_rate": 2.061073738537635e-05, | |
| "loss": 1.7901, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.13888270643222791, | |
| "grad_norm": 1.489214539527893, | |
| "learning_rate": 1.7527597583490822e-05, | |
| "loss": 1.8801, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.13888270643222791, | |
| "eval_loss": 2.036982297897339, | |
| "eval_runtime": 110.9189, | |
| "eval_samples_per_second": 4.264, | |
| "eval_steps_per_second": 2.137, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.1424438014689517, | |
| "grad_norm": 1.3451905250549316, | |
| "learning_rate": 1.4644660940672627e-05, | |
| "loss": 1.8144, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.1460048965056755, | |
| "grad_norm": 1.5614116191864014, | |
| "learning_rate": 1.1979701719998453e-05, | |
| "loss": 1.8939, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.1495659915423993, | |
| "grad_norm": 1.434024453163147, | |
| "learning_rate": 9.549150281252633e-06, | |
| "loss": 2.1714, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.15312708657912308, | |
| "grad_norm": 1.9399369955062866, | |
| "learning_rate": 7.367991782295391e-06, | |
| "loss": 2.3259, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.15668818161584688, | |
| "grad_norm": 1.6020128726959229, | |
| "learning_rate": 5.449673790581611e-06, | |
| "loss": 2.3047, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.16024927665257066, | |
| "grad_norm": 1.3612734079360962, | |
| "learning_rate": 3.8060233744356633e-06, | |
| "loss": 2.0302, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.16381037168929447, | |
| "grad_norm": 1.40798819065094, | |
| "learning_rate": 2.4471741852423237e-06, | |
| "loss": 1.9763, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.16737146672601824, | |
| "grad_norm": 1.7111047506332397, | |
| "learning_rate": 1.3815039801161721e-06, | |
| "loss": 2.0167, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.17093256176274205, | |
| "grad_norm": 1.519776701927185, | |
| "learning_rate": 6.15582970243117e-07, | |
| "loss": 2.0705, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.17449365679946582, | |
| "grad_norm": 1.4649944305419922, | |
| "learning_rate": 1.5413331334360182e-07, | |
| "loss": 1.8096, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.17805475183618963, | |
| "grad_norm": 1.449601173400879, | |
| "learning_rate": 0.0, | |
| "loss": 1.6536, | |
| "step": 50 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 50, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 13, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 5.08957047128064e+17, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |