| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.037993920972644375, | |
| "eval_steps": 5, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0007598784194528875, | |
| "grad_norm": 0.7123318910598755, | |
| "learning_rate": 1e-05, | |
| "loss": 22.2421, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0007598784194528875, | |
| "eval_loss": 11.113731384277344, | |
| "eval_runtime": 9.5997, | |
| "eval_samples_per_second": 115.42, | |
| "eval_steps_per_second": 14.48, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.001519756838905775, | |
| "grad_norm": 0.7559512257575989, | |
| "learning_rate": 2e-05, | |
| "loss": 22.2073, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.0022796352583586625, | |
| "grad_norm": 0.6938263177871704, | |
| "learning_rate": 3e-05, | |
| "loss": 22.2088, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.00303951367781155, | |
| "grad_norm": 0.7358536720275879, | |
| "learning_rate": 4e-05, | |
| "loss": 22.2193, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.003799392097264438, | |
| "grad_norm": 0.7246705293655396, | |
| "learning_rate": 5e-05, | |
| "loss": 22.2103, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.003799392097264438, | |
| "eval_loss": 11.111109733581543, | |
| "eval_runtime": 9.1747, | |
| "eval_samples_per_second": 120.767, | |
| "eval_steps_per_second": 15.15, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.004559270516717325, | |
| "grad_norm": 0.7420904636383057, | |
| "learning_rate": 6e-05, | |
| "loss": 22.2077, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.005319148936170213, | |
| "grad_norm": 0.7143452763557434, | |
| "learning_rate": 7e-05, | |
| "loss": 22.2248, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.0060790273556231, | |
| "grad_norm": 0.7711450457572937, | |
| "learning_rate": 8e-05, | |
| "loss": 22.2106, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.006838905775075988, | |
| "grad_norm": 0.7395123243331909, | |
| "learning_rate": 9e-05, | |
| "loss": 22.2055, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.007598784194528876, | |
| "grad_norm": 0.724628746509552, | |
| "learning_rate": 0.0001, | |
| "loss": 22.2055, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.007598784194528876, | |
| "eval_loss": 11.098109245300293, | |
| "eval_runtime": 9.6114, | |
| "eval_samples_per_second": 115.28, | |
| "eval_steps_per_second": 14.462, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.008358662613981762, | |
| "grad_norm": 0.7743872404098511, | |
| "learning_rate": 9.98458666866564e-05, | |
| "loss": 22.1964, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.00911854103343465, | |
| "grad_norm": 0.8174950480461121, | |
| "learning_rate": 9.938441702975689e-05, | |
| "loss": 22.1942, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.009878419452887538, | |
| "grad_norm": 0.8450507521629333, | |
| "learning_rate": 9.861849601988383e-05, | |
| "loss": 22.1764, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.010638297872340425, | |
| "grad_norm": 0.7814104557037354, | |
| "learning_rate": 9.755282581475769e-05, | |
| "loss": 22.1768, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.011398176291793313, | |
| "grad_norm": 0.8507615923881531, | |
| "learning_rate": 9.619397662556435e-05, | |
| "loss": 22.1678, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.011398176291793313, | |
| "eval_loss": 11.078285217285156, | |
| "eval_runtime": 9.4456, | |
| "eval_samples_per_second": 117.304, | |
| "eval_steps_per_second": 14.716, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.0121580547112462, | |
| "grad_norm": 1.170660138130188, | |
| "learning_rate": 9.45503262094184e-05, | |
| "loss": 22.2144, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.012917933130699088, | |
| "grad_norm": 1.2283384799957275, | |
| "learning_rate": 9.263200821770461e-05, | |
| "loss": 22.2188, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.013677811550151976, | |
| "grad_norm": 0.8715210556983948, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 22.1407, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.014437689969604863, | |
| "grad_norm": 0.8453183174133301, | |
| "learning_rate": 8.802029828000156e-05, | |
| "loss": 22.1368, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.015197568389057751, | |
| "grad_norm": 0.8434128761291504, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 22.1207, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.015197568389057751, | |
| "eval_loss": 11.058117866516113, | |
| "eval_runtime": 8.7467, | |
| "eval_samples_per_second": 126.676, | |
| "eval_steps_per_second": 15.892, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.015957446808510637, | |
| "grad_norm": 0.9485874176025391, | |
| "learning_rate": 8.247240241650918e-05, | |
| "loss": 22.0967, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.016717325227963525, | |
| "grad_norm": 0.9110122323036194, | |
| "learning_rate": 7.938926261462366e-05, | |
| "loss": 22.0956, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.017477203647416412, | |
| "grad_norm": 0.8467977046966553, | |
| "learning_rate": 7.612492823579745e-05, | |
| "loss": 22.1266, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.0182370820668693, | |
| "grad_norm": 0.8551048636436462, | |
| "learning_rate": 7.269952498697734e-05, | |
| "loss": 22.1062, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.018996960486322188, | |
| "grad_norm": 0.8727412819862366, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 22.0855, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.018996960486322188, | |
| "eval_loss": 11.037599563598633, | |
| "eval_runtime": 9.5705, | |
| "eval_samples_per_second": 115.772, | |
| "eval_steps_per_second": 14.524, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.019756838905775075, | |
| "grad_norm": 0.966887891292572, | |
| "learning_rate": 6.545084971874738e-05, | |
| "loss": 22.063, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.020516717325227963, | |
| "grad_norm": 1.0077238082885742, | |
| "learning_rate": 6.167226819279528e-05, | |
| "loss": 22.0584, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.02127659574468085, | |
| "grad_norm": 0.9766594171524048, | |
| "learning_rate": 5.782172325201155e-05, | |
| "loss": 22.0625, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.022036474164133738, | |
| "grad_norm": 0.9364005327224731, | |
| "learning_rate": 5.392295478639225e-05, | |
| "loss": 22.0474, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.022796352583586626, | |
| "grad_norm": 1.0432188510894775, | |
| "learning_rate": 5e-05, | |
| "loss": 22.0193, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.022796352583586626, | |
| "eval_loss": 11.019878387451172, | |
| "eval_runtime": 9.1273, | |
| "eval_samples_per_second": 121.394, | |
| "eval_steps_per_second": 15.229, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.023556231003039513, | |
| "grad_norm": 0.990106463432312, | |
| "learning_rate": 4.607704521360776e-05, | |
| "loss": 22.0551, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.0243161094224924, | |
| "grad_norm": 0.9892701506614685, | |
| "learning_rate": 4.2178276747988446e-05, | |
| "loss": 22.0255, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.02507598784194529, | |
| "grad_norm": 1.0256024599075317, | |
| "learning_rate": 3.832773180720475e-05, | |
| "loss": 22.0296, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.025835866261398176, | |
| "grad_norm": 0.9178345203399658, | |
| "learning_rate": 3.4549150281252636e-05, | |
| "loss": 22.0515, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.026595744680851064, | |
| "grad_norm": 1.0213465690612793, | |
| "learning_rate": 3.086582838174551e-05, | |
| "loss": 22.046, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.026595744680851064, | |
| "eval_loss": 11.006331443786621, | |
| "eval_runtime": 9.6408, | |
| "eval_samples_per_second": 114.929, | |
| "eval_steps_per_second": 14.418, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.02735562310030395, | |
| "grad_norm": 1.0454298257827759, | |
| "learning_rate": 2.7300475013022663e-05, | |
| "loss": 22.0217, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.02811550151975684, | |
| "grad_norm": 1.0446940660476685, | |
| "learning_rate": 2.3875071764202563e-05, | |
| "loss": 21.9847, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.028875379939209727, | |
| "grad_norm": 1.061407446861267, | |
| "learning_rate": 2.061073738537635e-05, | |
| "loss": 22.0118, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.029635258358662615, | |
| "grad_norm": 1.0344719886779785, | |
| "learning_rate": 1.7527597583490822e-05, | |
| "loss": 21.9994, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.030395136778115502, | |
| "grad_norm": 1.0409643650054932, | |
| "learning_rate": 1.4644660940672627e-05, | |
| "loss": 22.0019, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.030395136778115502, | |
| "eval_loss": 10.998189926147461, | |
| "eval_runtime": 9.3925, | |
| "eval_samples_per_second": 117.967, | |
| "eval_steps_per_second": 14.799, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.03115501519756839, | |
| "grad_norm": 1.0946383476257324, | |
| "learning_rate": 1.1979701719998453e-05, | |
| "loss": 21.9642, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.031914893617021274, | |
| "grad_norm": 1.0578166246414185, | |
| "learning_rate": 9.549150281252633e-06, | |
| "loss": 21.9906, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.03267477203647416, | |
| "grad_norm": 1.0239158868789673, | |
| "learning_rate": 7.367991782295391e-06, | |
| "loss": 21.9779, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.03343465045592705, | |
| "grad_norm": 1.0567377805709839, | |
| "learning_rate": 5.449673790581611e-06, | |
| "loss": 21.9981, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.03419452887537994, | |
| "grad_norm": 1.0055408477783203, | |
| "learning_rate": 3.8060233744356633e-06, | |
| "loss": 22.0115, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.03419452887537994, | |
| "eval_loss": 10.994943618774414, | |
| "eval_runtime": 9.5858, | |
| "eval_samples_per_second": 115.587, | |
| "eval_steps_per_second": 14.501, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.034954407294832825, | |
| "grad_norm": 1.1111423969268799, | |
| "learning_rate": 2.4471741852423237e-06, | |
| "loss": 21.9687, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.03571428571428571, | |
| "grad_norm": 1.0502803325653076, | |
| "learning_rate": 1.3815039801161721e-06, | |
| "loss": 21.985, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.0364741641337386, | |
| "grad_norm": 0.9838832020759583, | |
| "learning_rate": 6.15582970243117e-07, | |
| "loss": 21.9997, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.03723404255319149, | |
| "grad_norm": 1.068258285522461, | |
| "learning_rate": 1.5413331334360182e-07, | |
| "loss": 21.9884, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.037993920972644375, | |
| "grad_norm": 1.0668437480926514, | |
| "learning_rate": 0.0, | |
| "loss": 21.9595, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.037993920972644375, | |
| "eval_loss": 10.994412422180176, | |
| "eval_runtime": 9.4009, | |
| "eval_samples_per_second": 117.861, | |
| "eval_steps_per_second": 14.786, | |
| "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": 2121554460672.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |