| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.005623031938821413, | |
| "eval_steps": 5, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.00011246063877642825, | |
| "grad_norm": 0.5725249648094177, | |
| "learning_rate": 1e-05, | |
| "loss": 1.5136, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.00011246063877642825, | |
| "eval_loss": 1.4840327501296997, | |
| "eval_runtime": 825.6591, | |
| "eval_samples_per_second": 9.069, | |
| "eval_steps_per_second": 1.134, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0002249212775528565, | |
| "grad_norm": 0.5346225500106812, | |
| "learning_rate": 2e-05, | |
| "loss": 1.39, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.00033738191632928474, | |
| "grad_norm": 0.5900377035140991, | |
| "learning_rate": 3e-05, | |
| "loss": 1.5405, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.000449842555105713, | |
| "grad_norm": 0.5621856451034546, | |
| "learning_rate": 4e-05, | |
| "loss": 1.5007, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.0005623031938821412, | |
| "grad_norm": 0.7087019085884094, | |
| "learning_rate": 5e-05, | |
| "loss": 1.4067, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.0005623031938821412, | |
| "eval_loss": 1.460050106048584, | |
| "eval_runtime": 827.0883, | |
| "eval_samples_per_second": 9.053, | |
| "eval_steps_per_second": 1.132, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.0006747638326585695, | |
| "grad_norm": 0.6442967653274536, | |
| "learning_rate": 6e-05, | |
| "loss": 1.4906, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.0007872244714349977, | |
| "grad_norm": 0.7573694586753845, | |
| "learning_rate": 7e-05, | |
| "loss": 1.4362, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.000899685110211426, | |
| "grad_norm": 0.8192639946937561, | |
| "learning_rate": 8e-05, | |
| "loss": 1.4622, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.0010121457489878543, | |
| "grad_norm": 0.6494068503379822, | |
| "learning_rate": 9e-05, | |
| "loss": 1.3181, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0011246063877642825, | |
| "grad_norm": 0.6658048033714294, | |
| "learning_rate": 0.0001, | |
| "loss": 1.284, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.0011246063877642825, | |
| "eval_loss": 1.2558841705322266, | |
| "eval_runtime": 827.2292, | |
| "eval_samples_per_second": 9.052, | |
| "eval_steps_per_second": 1.131, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.0012370670265407107, | |
| "grad_norm": 0.5474720597267151, | |
| "learning_rate": 9.98458666866564e-05, | |
| "loss": 1.148, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.001349527665317139, | |
| "grad_norm": 0.6393503546714783, | |
| "learning_rate": 9.938441702975689e-05, | |
| "loss": 1.1735, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.0014619883040935672, | |
| "grad_norm": 0.840015709400177, | |
| "learning_rate": 9.861849601988383e-05, | |
| "loss": 1.1667, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.0015744489428699954, | |
| "grad_norm": 0.7437843680381775, | |
| "learning_rate": 9.755282581475769e-05, | |
| "loss": 1.188, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.0016869095816464238, | |
| "grad_norm": 0.5781041979789734, | |
| "learning_rate": 9.619397662556435e-05, | |
| "loss": 1.1843, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.0016869095816464238, | |
| "eval_loss": 1.1295948028564453, | |
| "eval_runtime": 826.7278, | |
| "eval_samples_per_second": 9.057, | |
| "eval_steps_per_second": 1.132, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.001799370220422852, | |
| "grad_norm": 0.5727628469467163, | |
| "learning_rate": 9.45503262094184e-05, | |
| "loss": 1.0947, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.0019118308591992803, | |
| "grad_norm": 0.645592987537384, | |
| "learning_rate": 9.263200821770461e-05, | |
| "loss": 1.1366, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.0020242914979757085, | |
| "grad_norm": 0.5597833395004272, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 1.0325, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.002136752136752137, | |
| "grad_norm": 0.5782056450843811, | |
| "learning_rate": 8.802029828000156e-05, | |
| "loss": 1.096, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.002249212775528565, | |
| "grad_norm": 0.5130562782287598, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 1.1302, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.002249212775528565, | |
| "eval_loss": 1.069796085357666, | |
| "eval_runtime": 826.9105, | |
| "eval_samples_per_second": 9.055, | |
| "eval_steps_per_second": 1.132, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.0023616734143049934, | |
| "grad_norm": 0.4995030462741852, | |
| "learning_rate": 8.247240241650918e-05, | |
| "loss": 1.0282, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.0024741340530814214, | |
| "grad_norm": 0.5298396944999695, | |
| "learning_rate": 7.938926261462366e-05, | |
| "loss": 1.0888, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.00258659469185785, | |
| "grad_norm": 0.513755202293396, | |
| "learning_rate": 7.612492823579745e-05, | |
| "loss": 1.0107, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.002699055330634278, | |
| "grad_norm": 0.5589742064476013, | |
| "learning_rate": 7.269952498697734e-05, | |
| "loss": 1.0702, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.0028115159694107063, | |
| "grad_norm": 0.5653223991394043, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 1.0512, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.0028115159694107063, | |
| "eval_loss": 1.0359995365142822, | |
| "eval_runtime": 826.9292, | |
| "eval_samples_per_second": 9.055, | |
| "eval_steps_per_second": 1.132, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.0029239766081871343, | |
| "grad_norm": 0.5085208415985107, | |
| "learning_rate": 6.545084971874738e-05, | |
| "loss": 0.9733, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.003036437246963563, | |
| "grad_norm": 0.5659909248352051, | |
| "learning_rate": 6.167226819279528e-05, | |
| "loss": 1.0632, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.003148897885739991, | |
| "grad_norm": 0.5472968816757202, | |
| "learning_rate": 5.782172325201155e-05, | |
| "loss": 1.0175, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.0032613585245164192, | |
| "grad_norm": 0.524185299873352, | |
| "learning_rate": 5.392295478639225e-05, | |
| "loss": 1.0288, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.0033738191632928477, | |
| "grad_norm": 0.5419555306434631, | |
| "learning_rate": 5e-05, | |
| "loss": 1.0801, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.0033738191632928477, | |
| "eval_loss": 1.0140562057495117, | |
| "eval_runtime": 826.2541, | |
| "eval_samples_per_second": 9.063, | |
| "eval_steps_per_second": 1.133, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.0034862798020692757, | |
| "grad_norm": 0.44709768891334534, | |
| "learning_rate": 4.607704521360776e-05, | |
| "loss": 0.8995, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.003598740440845704, | |
| "grad_norm": 0.45280900597572327, | |
| "learning_rate": 4.2178276747988446e-05, | |
| "loss": 1.0176, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.003711201079622132, | |
| "grad_norm": 0.4053972661495209, | |
| "learning_rate": 3.832773180720475e-05, | |
| "loss": 0.9466, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.0038236617183985606, | |
| "grad_norm": 0.4766600430011749, | |
| "learning_rate": 3.4549150281252636e-05, | |
| "loss": 1.0009, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.003936122357174989, | |
| "grad_norm": 0.45294103026390076, | |
| "learning_rate": 3.086582838174551e-05, | |
| "loss": 1.0124, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.003936122357174989, | |
| "eval_loss": 1.0017218589782715, | |
| "eval_runtime": 827.7032, | |
| "eval_samples_per_second": 9.047, | |
| "eval_steps_per_second": 1.131, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.004048582995951417, | |
| "grad_norm": 0.47154468297958374, | |
| "learning_rate": 2.7300475013022663e-05, | |
| "loss": 0.9283, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.0041610436347278455, | |
| "grad_norm": 0.46065637469291687, | |
| "learning_rate": 2.3875071764202563e-05, | |
| "loss": 0.9644, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.004273504273504274, | |
| "grad_norm": 0.43747445940971375, | |
| "learning_rate": 2.061073738537635e-05, | |
| "loss": 0.9395, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.0043859649122807015, | |
| "grad_norm": 0.4837368428707123, | |
| "learning_rate": 1.7527597583490822e-05, | |
| "loss": 0.9991, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.00449842555105713, | |
| "grad_norm": 0.43076515197753906, | |
| "learning_rate": 1.4644660940672627e-05, | |
| "loss": 0.9176, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.00449842555105713, | |
| "eval_loss": 0.9943129420280457, | |
| "eval_runtime": 826.2381, | |
| "eval_samples_per_second": 9.063, | |
| "eval_steps_per_second": 1.133, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.004610886189833558, | |
| "grad_norm": 0.5260198712348938, | |
| "learning_rate": 1.1979701719998453e-05, | |
| "loss": 1.0163, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.004723346828609987, | |
| "grad_norm": 0.4770829677581787, | |
| "learning_rate": 9.549150281252633e-06, | |
| "loss": 1.0087, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.004835807467386414, | |
| "grad_norm": 0.5031581521034241, | |
| "learning_rate": 7.367991782295391e-06, | |
| "loss": 1.0117, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.004948268106162843, | |
| "grad_norm": 0.5243813395500183, | |
| "learning_rate": 5.449673790581611e-06, | |
| "loss": 1.1394, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.005060728744939271, | |
| "grad_norm": 0.49357548356056213, | |
| "learning_rate": 3.8060233744356633e-06, | |
| "loss": 1.002, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.005060728744939271, | |
| "eval_loss": 0.9909366965293884, | |
| "eval_runtime": 825.9423, | |
| "eval_samples_per_second": 9.066, | |
| "eval_steps_per_second": 1.133, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.0051731893837157, | |
| "grad_norm": 0.46298205852508545, | |
| "learning_rate": 2.4471741852423237e-06, | |
| "loss": 0.9586, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.005285650022492127, | |
| "grad_norm": 0.4490233063697815, | |
| "learning_rate": 1.3815039801161721e-06, | |
| "loss": 0.9514, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.005398110661268556, | |
| "grad_norm": 0.44175946712493896, | |
| "learning_rate": 6.15582970243117e-07, | |
| "loss": 0.8795, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.005510571300044984, | |
| "grad_norm": 0.4616028368473053, | |
| "learning_rate": 1.5413331334360182e-07, | |
| "loss": 0.9672, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.005623031938821413, | |
| "grad_norm": 0.456407368183136, | |
| "learning_rate": 0.0, | |
| "loss": 0.9266, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.005623031938821413, | |
| "eval_loss": 0.9902380108833313, | |
| "eval_runtime": 826.0197, | |
| "eval_samples_per_second": 9.065, | |
| "eval_steps_per_second": 1.133, | |
| "step": 50 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 50, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 25, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 3.70943641780224e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |