| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.1711229946524064, | |
| "eval_steps": 13, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.003422459893048128, | |
| "grad_norm": 1.7999919652938843, | |
| "learning_rate": 1e-05, | |
| "loss": 1.43, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.003422459893048128, | |
| "eval_loss": 1.5696145296096802, | |
| "eval_runtime": 93.9699, | |
| "eval_samples_per_second": 5.246, | |
| "eval_steps_per_second": 2.629, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.006844919786096256, | |
| "grad_norm": 1.4601882696151733, | |
| "learning_rate": 2e-05, | |
| "loss": 1.5299, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.010267379679144385, | |
| "grad_norm": 1.357947587966919, | |
| "learning_rate": 3e-05, | |
| "loss": 1.4347, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.013689839572192513, | |
| "grad_norm": 1.7744206190109253, | |
| "learning_rate": 4e-05, | |
| "loss": 1.5207, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.017112299465240642, | |
| "grad_norm": 1.8218324184417725, | |
| "learning_rate": 5e-05, | |
| "loss": 1.6008, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.02053475935828877, | |
| "grad_norm": 1.1797770261764526, | |
| "learning_rate": 6e-05, | |
| "loss": 1.2966, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.023957219251336898, | |
| "grad_norm": 1.1083546876907349, | |
| "learning_rate": 7e-05, | |
| "loss": 1.2069, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.027379679144385025, | |
| "grad_norm": 0.9623205661773682, | |
| "learning_rate": 8e-05, | |
| "loss": 1.1496, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.030802139037433156, | |
| "grad_norm": 0.78102046251297, | |
| "learning_rate": 9e-05, | |
| "loss": 1.3235, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.034224598930481284, | |
| "grad_norm": 0.8298507332801819, | |
| "learning_rate": 0.0001, | |
| "loss": 1.212, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.03764705882352941, | |
| "grad_norm": 0.8449644446372986, | |
| "learning_rate": 9.98458666866564e-05, | |
| "loss": 1.0372, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.04106951871657754, | |
| "grad_norm": 0.8819606900215149, | |
| "learning_rate": 9.938441702975689e-05, | |
| "loss": 1.1813, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.04449197860962567, | |
| "grad_norm": 0.8829816579818726, | |
| "learning_rate": 9.861849601988383e-05, | |
| "loss": 0.9319, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.04449197860962567, | |
| "eval_loss": 1.018886923789978, | |
| "eval_runtime": 17.4921, | |
| "eval_samples_per_second": 28.184, | |
| "eval_steps_per_second": 14.121, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.047914438502673795, | |
| "grad_norm": 0.8753724098205566, | |
| "learning_rate": 9.755282581475769e-05, | |
| "loss": 0.9516, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.051336898395721926, | |
| "grad_norm": 0.8921568989753723, | |
| "learning_rate": 9.619397662556435e-05, | |
| "loss": 0.8387, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.05475935828877005, | |
| "grad_norm": 0.8636670708656311, | |
| "learning_rate": 9.45503262094184e-05, | |
| "loss": 0.8335, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.05818181818181818, | |
| "grad_norm": 0.7787020802497864, | |
| "learning_rate": 9.263200821770461e-05, | |
| "loss": 0.8278, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.06160427807486631, | |
| "grad_norm": 0.9087046980857849, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 0.8099, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.06502673796791444, | |
| "grad_norm": 0.7061587572097778, | |
| "learning_rate": 8.802029828000156e-05, | |
| "loss": 0.6824, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.06844919786096257, | |
| "grad_norm": 0.8060164451599121, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 0.7852, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.07187165775401069, | |
| "grad_norm": 0.8172853589057922, | |
| "learning_rate": 8.247240241650918e-05, | |
| "loss": 0.6113, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.07529411764705882, | |
| "grad_norm": 1.1956216096878052, | |
| "learning_rate": 7.938926261462366e-05, | |
| "loss": 0.959, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.07871657754010696, | |
| "grad_norm": 0.8757884502410889, | |
| "learning_rate": 7.612492823579745e-05, | |
| "loss": 0.8671, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.08213903743315508, | |
| "grad_norm": 0.8951571583747864, | |
| "learning_rate": 7.269952498697734e-05, | |
| "loss": 0.5991, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.0855614973262032, | |
| "grad_norm": 1.3083561658859253, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 0.9128, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.08898395721925134, | |
| "grad_norm": 0.9369482398033142, | |
| "learning_rate": 6.545084971874738e-05, | |
| "loss": 0.5134, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.08898395721925134, | |
| "eval_loss": 0.6822815537452698, | |
| "eval_runtime": 17.4703, | |
| "eval_samples_per_second": 28.219, | |
| "eval_steps_per_second": 14.138, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.09240641711229947, | |
| "grad_norm": 0.732320249080658, | |
| "learning_rate": 6.167226819279528e-05, | |
| "loss": 0.5086, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.09582887700534759, | |
| "grad_norm": 0.6814180016517639, | |
| "learning_rate": 5.782172325201155e-05, | |
| "loss": 0.5471, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.09925133689839573, | |
| "grad_norm": 0.7060202956199646, | |
| "learning_rate": 5.392295478639225e-05, | |
| "loss": 0.3652, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.10267379679144385, | |
| "grad_norm": 0.9559823274612427, | |
| "learning_rate": 5e-05, | |
| "loss": 0.8032, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.10609625668449198, | |
| "grad_norm": 1.1005346775054932, | |
| "learning_rate": 4.607704521360776e-05, | |
| "loss": 0.8232, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.1095187165775401, | |
| "grad_norm": 0.7558871507644653, | |
| "learning_rate": 4.2178276747988446e-05, | |
| "loss": 0.5147, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.11294117647058824, | |
| "grad_norm": 0.6404812932014465, | |
| "learning_rate": 3.832773180720475e-05, | |
| "loss": 0.4, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.11636363636363636, | |
| "grad_norm": 0.8565809726715088, | |
| "learning_rate": 3.4549150281252636e-05, | |
| "loss": 0.5938, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.11978609625668449, | |
| "grad_norm": 0.6082307696342468, | |
| "learning_rate": 3.086582838174551e-05, | |
| "loss": 0.4025, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.12320855614973263, | |
| "grad_norm": 0.7612841725349426, | |
| "learning_rate": 2.7300475013022663e-05, | |
| "loss": 0.5164, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.12663101604278074, | |
| "grad_norm": 0.7494601011276245, | |
| "learning_rate": 2.3875071764202563e-05, | |
| "loss": 0.442, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.1300534759358289, | |
| "grad_norm": 0.6392002701759338, | |
| "learning_rate": 2.061073738537635e-05, | |
| "loss": 0.2912, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.133475935828877, | |
| "grad_norm": 0.7933446168899536, | |
| "learning_rate": 1.7527597583490822e-05, | |
| "loss": 0.3706, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.133475935828877, | |
| "eval_loss": 0.5577317476272583, | |
| "eval_runtime": 17.4945, | |
| "eval_samples_per_second": 28.18, | |
| "eval_steps_per_second": 14.119, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.13689839572192514, | |
| "grad_norm": 0.8070633411407471, | |
| "learning_rate": 1.4644660940672627e-05, | |
| "loss": 0.5244, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.14032085561497326, | |
| "grad_norm": 0.762014627456665, | |
| "learning_rate": 1.1979701719998453e-05, | |
| "loss": 0.4261, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.14374331550802139, | |
| "grad_norm": 0.5508111715316772, | |
| "learning_rate": 9.549150281252633e-06, | |
| "loss": 0.3714, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.1471657754010695, | |
| "grad_norm": 0.735578715801239, | |
| "learning_rate": 7.367991782295391e-06, | |
| "loss": 0.4624, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.15058823529411763, | |
| "grad_norm": 0.5103933811187744, | |
| "learning_rate": 5.449673790581611e-06, | |
| "loss": 0.244, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.15401069518716579, | |
| "grad_norm": 0.5898608565330505, | |
| "learning_rate": 3.8060233744356633e-06, | |
| "loss": 0.3896, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.1574331550802139, | |
| "grad_norm": 0.676190197467804, | |
| "learning_rate": 2.4471741852423237e-06, | |
| "loss": 0.3595, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.16085561497326203, | |
| "grad_norm": 0.5928998589515686, | |
| "learning_rate": 1.3815039801161721e-06, | |
| "loss": 0.4371, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.16427807486631016, | |
| "grad_norm": 0.815858781337738, | |
| "learning_rate": 6.15582970243117e-07, | |
| "loss": 0.413, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.16770053475935828, | |
| "grad_norm": 0.5120140910148621, | |
| "learning_rate": 1.5413331334360182e-07, | |
| "loss": 0.1915, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.1711229946524064, | |
| "grad_norm": 0.5748466849327087, | |
| "learning_rate": 0.0, | |
| "loss": 0.3791, | |
| "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": 1.4419086016512e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |