| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.03904724716907458, | |
| "eval_steps": 13, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0007809449433814916, | |
| "grad_norm": 175.4751434326172, | |
| "learning_rate": 1e-05, | |
| "loss": 36.1201, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0007809449433814916, | |
| "eval_loss": 3.8799195289611816, | |
| "eval_runtime": 315.5656, | |
| "eval_samples_per_second": 6.835, | |
| "eval_steps_per_second": 3.419, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0015618898867629833, | |
| "grad_norm": 238.4190673828125, | |
| "learning_rate": 2e-05, | |
| "loss": 50.8572, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.0023428348301444747, | |
| "grad_norm": 363.2473449707031, | |
| "learning_rate": 3e-05, | |
| "loss": 69.5408, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0031237797735259665, | |
| "grad_norm": 166.27882385253906, | |
| "learning_rate": 4e-05, | |
| "loss": 36.4834, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.003904724716907458, | |
| "grad_norm": 206.5617218017578, | |
| "learning_rate": 5e-05, | |
| "loss": 43.0769, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.004685669660288949, | |
| "grad_norm": 192.99514770507812, | |
| "learning_rate": 6e-05, | |
| "loss": 57.0914, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.005466614603670442, | |
| "grad_norm": 187.74777221679688, | |
| "learning_rate": 7e-05, | |
| "loss": 41.0429, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.006247559547051933, | |
| "grad_norm": 242.54119873046875, | |
| "learning_rate": 8e-05, | |
| "loss": 50.7796, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.007028504490433424, | |
| "grad_norm": 88.46190643310547, | |
| "learning_rate": 9e-05, | |
| "loss": 23.8709, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.007809449433814916, | |
| "grad_norm": 102.14177703857422, | |
| "learning_rate": 0.0001, | |
| "loss": 22.4249, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.008590394377196407, | |
| "grad_norm": 71.25434875488281, | |
| "learning_rate": 9.98458666866564e-05, | |
| "loss": 20.7456, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.009371339320577899, | |
| "grad_norm": 97.2330551147461, | |
| "learning_rate": 9.938441702975689e-05, | |
| "loss": 20.3148, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.01015228426395939, | |
| "grad_norm": 71.13471221923828, | |
| "learning_rate": 9.861849601988383e-05, | |
| "loss": 17.7955, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.01015228426395939, | |
| "eval_loss": 1.2057011127471924, | |
| "eval_runtime": 300.8171, | |
| "eval_samples_per_second": 7.17, | |
| "eval_steps_per_second": 3.587, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.010933229207340883, | |
| "grad_norm": 70.08533477783203, | |
| "learning_rate": 9.755282581475769e-05, | |
| "loss": 17.0262, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.011714174150722375, | |
| "grad_norm": 69.94759368896484, | |
| "learning_rate": 9.619397662556435e-05, | |
| "loss": 18.1821, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.012495119094103866, | |
| "grad_norm": 58.61181640625, | |
| "learning_rate": 9.45503262094184e-05, | |
| "loss": 17.1924, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.013276064037485357, | |
| "grad_norm": 54.452816009521484, | |
| "learning_rate": 9.263200821770461e-05, | |
| "loss": 14.6744, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.014057008980866849, | |
| "grad_norm": 67.27265930175781, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 17.592, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.01483795392424834, | |
| "grad_norm": 51.187679290771484, | |
| "learning_rate": 8.802029828000156e-05, | |
| "loss": 20.2565, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.015618898867629832, | |
| "grad_norm": 45.95616912841797, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 16.9602, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.016399843811011325, | |
| "grad_norm": 46.94184112548828, | |
| "learning_rate": 8.247240241650918e-05, | |
| "loss": 16.319, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.017180788754392814, | |
| "grad_norm": 43.87798309326172, | |
| "learning_rate": 7.938926261462366e-05, | |
| "loss": 12.6792, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.017961733697774308, | |
| "grad_norm": 51.43169403076172, | |
| "learning_rate": 7.612492823579745e-05, | |
| "loss": 16.1705, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.018742678641155797, | |
| "grad_norm": 47.817901611328125, | |
| "learning_rate": 7.269952498697734e-05, | |
| "loss": 11.841, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.01952362358453729, | |
| "grad_norm": 56.87557601928711, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 17.5764, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.02030456852791878, | |
| "grad_norm": 45.88191223144531, | |
| "learning_rate": 6.545084971874738e-05, | |
| "loss": 13.4572, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.02030456852791878, | |
| "eval_loss": 0.8506395816802979, | |
| "eval_runtime": 300.4403, | |
| "eval_samples_per_second": 7.179, | |
| "eval_steps_per_second": 3.591, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.021085513471300273, | |
| "grad_norm": 44.15971374511719, | |
| "learning_rate": 6.167226819279528e-05, | |
| "loss": 12.4669, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.021866458414681766, | |
| "grad_norm": 70.013916015625, | |
| "learning_rate": 5.782172325201155e-05, | |
| "loss": 12.9177, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.022647403358063256, | |
| "grad_norm": 59.45874786376953, | |
| "learning_rate": 5.392295478639225e-05, | |
| "loss": 13.9473, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.02342834830144475, | |
| "grad_norm": 49.43613815307617, | |
| "learning_rate": 5e-05, | |
| "loss": 13.1336, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.02420929324482624, | |
| "grad_norm": 51.60984420776367, | |
| "learning_rate": 4.607704521360776e-05, | |
| "loss": 11.6511, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.024990238188207732, | |
| "grad_norm": 39.98220443725586, | |
| "learning_rate": 4.2178276747988446e-05, | |
| "loss": 11.9204, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.025771183131589222, | |
| "grad_norm": 53.87978744506836, | |
| "learning_rate": 3.832773180720475e-05, | |
| "loss": 14.3878, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.026552128074970715, | |
| "grad_norm": 55.93902587890625, | |
| "learning_rate": 3.4549150281252636e-05, | |
| "loss": 10.2647, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.027333073018352205, | |
| "grad_norm": 45.10630798339844, | |
| "learning_rate": 3.086582838174551e-05, | |
| "loss": 12.807, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.028114017961733698, | |
| "grad_norm": 57.13920593261719, | |
| "learning_rate": 2.7300475013022663e-05, | |
| "loss": 14.7782, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.02889496290511519, | |
| "grad_norm": 39.96442413330078, | |
| "learning_rate": 2.3875071764202563e-05, | |
| "loss": 13.0561, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.02967590784849668, | |
| "grad_norm": 38.73209762573242, | |
| "learning_rate": 2.061073738537635e-05, | |
| "loss": 11.2156, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.030456852791878174, | |
| "grad_norm": 49.080196380615234, | |
| "learning_rate": 1.7527597583490822e-05, | |
| "loss": 12.6033, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.030456852791878174, | |
| "eval_loss": 0.7845144271850586, | |
| "eval_runtime": 300.1758, | |
| "eval_samples_per_second": 7.186, | |
| "eval_steps_per_second": 3.595, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.031237797735259663, | |
| "grad_norm": 47.26570510864258, | |
| "learning_rate": 1.4644660940672627e-05, | |
| "loss": 12.213, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.03201874267864115, | |
| "grad_norm": 37.84298324584961, | |
| "learning_rate": 1.1979701719998453e-05, | |
| "loss": 12.9798, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.03279968762202265, | |
| "grad_norm": 37.72480773925781, | |
| "learning_rate": 9.549150281252633e-06, | |
| "loss": 10.6722, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.03358063256540414, | |
| "grad_norm": 44.485225677490234, | |
| "learning_rate": 7.367991782295391e-06, | |
| "loss": 12.6417, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.03436157750878563, | |
| "grad_norm": 36.807350158691406, | |
| "learning_rate": 5.449673790581611e-06, | |
| "loss": 11.1903, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.035142522452167126, | |
| "grad_norm": 34.367584228515625, | |
| "learning_rate": 3.8060233744356633e-06, | |
| "loss": 11.6601, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.035923467395548615, | |
| "grad_norm": 48.056583404541016, | |
| "learning_rate": 2.4471741852423237e-06, | |
| "loss": 13.341, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.036704412338930105, | |
| "grad_norm": 46.177330017089844, | |
| "learning_rate": 1.3815039801161721e-06, | |
| "loss": 13.4865, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.037485357282311595, | |
| "grad_norm": 54.51680374145508, | |
| "learning_rate": 6.15582970243117e-07, | |
| "loss": 12.0828, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.03826630222569309, | |
| "grad_norm": 43.835418701171875, | |
| "learning_rate": 1.5413331334360182e-07, | |
| "loss": 12.4233, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.03904724716907458, | |
| "grad_norm": 44.20928955078125, | |
| "learning_rate": 0.0, | |
| "loss": 13.5881, | |
| "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": 2.812517994725376e+17, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |