| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.12326656394453005, | |
| "eval_steps": 13, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.002465331278890601, | |
| "grad_norm": 26.539125442504883, | |
| "learning_rate": 1e-05, | |
| "loss": 3.6602, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.002465331278890601, | |
| "eval_loss": 3.617913246154785, | |
| "eval_runtime": 29.3583, | |
| "eval_samples_per_second": 93.091, | |
| "eval_steps_per_second": 11.649, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.004930662557781202, | |
| "grad_norm": 24.928752899169922, | |
| "learning_rate": 2e-05, | |
| "loss": 3.5913, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.007395993836671803, | |
| "grad_norm": 25.013662338256836, | |
| "learning_rate": 3e-05, | |
| "loss": 3.4352, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.009861325115562404, | |
| "grad_norm": 26.976316452026367, | |
| "learning_rate": 4e-05, | |
| "loss": 3.4006, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.012326656394453005, | |
| "grad_norm": 19.72706413269043, | |
| "learning_rate": 5e-05, | |
| "loss": 3.0162, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.014791987673343606, | |
| "grad_norm": 17.657188415527344, | |
| "learning_rate": 6e-05, | |
| "loss": 2.4475, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.017257318952234205, | |
| "grad_norm": 22.630268096923828, | |
| "learning_rate": 7e-05, | |
| "loss": 1.8466, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.019722650231124808, | |
| "grad_norm": 15.391894340515137, | |
| "learning_rate": 8e-05, | |
| "loss": 1.1958, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.022187981510015407, | |
| "grad_norm": 17.156049728393555, | |
| "learning_rate": 9e-05, | |
| "loss": 0.9063, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.02465331278890601, | |
| "grad_norm": 14.354949951171875, | |
| "learning_rate": 0.0001, | |
| "loss": 0.5376, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.02711864406779661, | |
| "grad_norm": 4.375965595245361, | |
| "learning_rate": 9.98458666866564e-05, | |
| "loss": 0.1246, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.029583975346687212, | |
| "grad_norm": 1.0381114482879639, | |
| "learning_rate": 9.938441702975689e-05, | |
| "loss": 0.0747, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.03204930662557781, | |
| "grad_norm": 2.4687139987945557, | |
| "learning_rate": 9.861849601988383e-05, | |
| "loss": 0.0975, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.03204930662557781, | |
| "eval_loss": 0.04013116657733917, | |
| "eval_runtime": 19.4505, | |
| "eval_samples_per_second": 140.511, | |
| "eval_steps_per_second": 17.583, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.03451463790446841, | |
| "grad_norm": 0.8402537107467651, | |
| "learning_rate": 9.755282581475769e-05, | |
| "loss": 0.0367, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.03697996918335902, | |
| "grad_norm": 0.7409534454345703, | |
| "learning_rate": 9.619397662556435e-05, | |
| "loss": 0.0414, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.039445300462249616, | |
| "grad_norm": 0.5837918519973755, | |
| "learning_rate": 9.45503262094184e-05, | |
| "loss": 0.0256, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.041910631741140215, | |
| "grad_norm": 2.8236846923828125, | |
| "learning_rate": 9.263200821770461e-05, | |
| "loss": 0.039, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.044375963020030815, | |
| "grad_norm": 0.8533737063407898, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 0.0401, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.04684129429892142, | |
| "grad_norm": 0.8649203181266785, | |
| "learning_rate": 8.802029828000156e-05, | |
| "loss": 0.0348, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.04930662557781202, | |
| "grad_norm": 0.9208324551582336, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 0.0302, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.05177195685670262, | |
| "grad_norm": 3.5557940006256104, | |
| "learning_rate": 8.247240241650918e-05, | |
| "loss": 0.0353, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.05423728813559322, | |
| "grad_norm": 1.0510553121566772, | |
| "learning_rate": 7.938926261462366e-05, | |
| "loss": 0.022, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.05670261941448382, | |
| "grad_norm": 0.9362481832504272, | |
| "learning_rate": 7.612492823579745e-05, | |
| "loss": 0.0215, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.059167950693374424, | |
| "grad_norm": 1.0250906944274902, | |
| "learning_rate": 7.269952498697734e-05, | |
| "loss": 0.0336, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.061633281972265024, | |
| "grad_norm": 0.12144595384597778, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 0.0256, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.06409861325115562, | |
| "grad_norm": 0.7493911981582642, | |
| "learning_rate": 6.545084971874738e-05, | |
| "loss": 0.0262, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.06409861325115562, | |
| "eval_loss": 0.023128492757678032, | |
| "eval_runtime": 19.4515, | |
| "eval_samples_per_second": 140.503, | |
| "eval_steps_per_second": 17.582, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.06656394453004623, | |
| "grad_norm": 0.946167528629303, | |
| "learning_rate": 6.167226819279528e-05, | |
| "loss": 0.0381, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.06902927580893682, | |
| "grad_norm": 1.160494089126587, | |
| "learning_rate": 5.782172325201155e-05, | |
| "loss": 0.0201, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.07149460708782743, | |
| "grad_norm": 0.37332940101623535, | |
| "learning_rate": 5.392295478639225e-05, | |
| "loss": 0.022, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.07395993836671803, | |
| "grad_norm": 0.8763566613197327, | |
| "learning_rate": 5e-05, | |
| "loss": 0.0371, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.07642526964560863, | |
| "grad_norm": 0.07144226133823395, | |
| "learning_rate": 4.607704521360776e-05, | |
| "loss": 0.0063, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.07889060092449923, | |
| "grad_norm": 0.2712821066379547, | |
| "learning_rate": 4.2178276747988446e-05, | |
| "loss": 0.0231, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.08135593220338982, | |
| "grad_norm": 0.79026198387146, | |
| "learning_rate": 3.832773180720475e-05, | |
| "loss": 0.014, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.08382126348228043, | |
| "grad_norm": 0.7974649667739868, | |
| "learning_rate": 3.4549150281252636e-05, | |
| "loss": 0.0096, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.08628659476117104, | |
| "grad_norm": 0.4885225296020508, | |
| "learning_rate": 3.086582838174551e-05, | |
| "loss": 0.0126, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.08875192604006163, | |
| "grad_norm": 0.5886369347572327, | |
| "learning_rate": 2.7300475013022663e-05, | |
| "loss": 0.0296, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.09121725731895224, | |
| "grad_norm": 1.1059995889663696, | |
| "learning_rate": 2.3875071764202563e-05, | |
| "loss": 0.0491, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.09368258859784284, | |
| "grad_norm": 2.0051498413085938, | |
| "learning_rate": 2.061073738537635e-05, | |
| "loss": 0.0301, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.09614791987673343, | |
| "grad_norm": 0.6557801365852356, | |
| "learning_rate": 1.7527597583490822e-05, | |
| "loss": 0.0304, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.09614791987673343, | |
| "eval_loss": 0.022078825160861015, | |
| "eval_runtime": 19.4482, | |
| "eval_samples_per_second": 140.527, | |
| "eval_steps_per_second": 17.585, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.09861325115562404, | |
| "grad_norm": 0.2994353771209717, | |
| "learning_rate": 1.4644660940672627e-05, | |
| "loss": 0.0195, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.10107858243451463, | |
| "grad_norm": 0.19609463214874268, | |
| "learning_rate": 1.1979701719998453e-05, | |
| "loss": 0.0397, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.10354391371340524, | |
| "grad_norm": 0.8093880414962769, | |
| "learning_rate": 9.549150281252633e-06, | |
| "loss": 0.0186, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.10600924499229585, | |
| "grad_norm": 0.6355484127998352, | |
| "learning_rate": 7.367991782295391e-06, | |
| "loss": 0.024, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.10847457627118644, | |
| "grad_norm": 0.41448116302490234, | |
| "learning_rate": 5.449673790581611e-06, | |
| "loss": 0.0117, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.11093990755007704, | |
| "grad_norm": 0.775438129901886, | |
| "learning_rate": 3.8060233744356633e-06, | |
| "loss": 0.0383, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.11340523882896764, | |
| "grad_norm": 0.8112552165985107, | |
| "learning_rate": 2.4471741852423237e-06, | |
| "loss": 0.0306, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.11587057010785824, | |
| "grad_norm": 0.46484261751174927, | |
| "learning_rate": 1.3815039801161721e-06, | |
| "loss": 0.0129, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.11833590138674885, | |
| "grad_norm": 0.7908012866973877, | |
| "learning_rate": 6.15582970243117e-07, | |
| "loss": 0.0282, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.12080123266563944, | |
| "grad_norm": 0.3184359073638916, | |
| "learning_rate": 1.5413331334360182e-07, | |
| "loss": 0.0163, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.12326656394453005, | |
| "grad_norm": 0.32925423979759216, | |
| "learning_rate": 0.0, | |
| "loss": 0.0087, | |
| "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": 5.7676344066048e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |