| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.11927836588638735, | |
| "eval_steps": 13, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0023855673177277472, | |
| "grad_norm": 1.3246102333068848, | |
| "learning_rate": 1e-05, | |
| "loss": 0.9265, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0023855673177277472, | |
| "eval_loss": 0.9870346188545227, | |
| "eval_runtime": 65.1117, | |
| "eval_samples_per_second": 10.843, | |
| "eval_steps_per_second": 5.421, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0047711346354554944, | |
| "grad_norm": 1.5241637229919434, | |
| "learning_rate": 2e-05, | |
| "loss": 1.0298, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.007156701953183241, | |
| "grad_norm": 1.3183329105377197, | |
| "learning_rate": 3e-05, | |
| "loss": 0.9629, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.009542269270910989, | |
| "grad_norm": 1.3316704034805298, | |
| "learning_rate": 4e-05, | |
| "loss": 0.9491, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.011927836588638736, | |
| "grad_norm": 1.2836781740188599, | |
| "learning_rate": 5e-05, | |
| "loss": 0.8294, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.014313403906366482, | |
| "grad_norm": 1.3446112871170044, | |
| "learning_rate": 6e-05, | |
| "loss": 0.7596, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.01669897122409423, | |
| "grad_norm": 1.3533251285552979, | |
| "learning_rate": 7e-05, | |
| "loss": 0.8652, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.019084538541821978, | |
| "grad_norm": 1.2886685132980347, | |
| "learning_rate": 8e-05, | |
| "loss": 0.6794, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.021470105859549723, | |
| "grad_norm": 1.0436530113220215, | |
| "learning_rate": 9e-05, | |
| "loss": 0.5633, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.02385567317727747, | |
| "grad_norm": 0.8520669937133789, | |
| "learning_rate": 0.0001, | |
| "loss": 0.4955, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.02624124049500522, | |
| "grad_norm": 0.7037628889083862, | |
| "learning_rate": 9.98458666866564e-05, | |
| "loss": 0.3973, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.028626807812732965, | |
| "grad_norm": 0.8181518912315369, | |
| "learning_rate": 9.938441702975689e-05, | |
| "loss": 0.4359, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.031012375130460713, | |
| "grad_norm": 0.6722888946533203, | |
| "learning_rate": 9.861849601988383e-05, | |
| "loss": 0.3231, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.031012375130460713, | |
| "eval_loss": 0.303219199180603, | |
| "eval_runtime": 16.6267, | |
| "eval_samples_per_second": 42.462, | |
| "eval_steps_per_second": 21.231, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.03339794244818846, | |
| "grad_norm": 0.6669241786003113, | |
| "learning_rate": 9.755282581475769e-05, | |
| "loss": 0.3008, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.035783509765916204, | |
| "grad_norm": 0.6451194882392883, | |
| "learning_rate": 9.619397662556435e-05, | |
| "loss": 0.2879, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.038169077083643956, | |
| "grad_norm": 0.6567999720573425, | |
| "learning_rate": 9.45503262094184e-05, | |
| "loss": 0.2355, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.0405546444013717, | |
| "grad_norm": 0.534218430519104, | |
| "learning_rate": 9.263200821770461e-05, | |
| "loss": 0.1906, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.042940211719099446, | |
| "grad_norm": 0.7300930023193359, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 0.2783, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.0453257790368272, | |
| "grad_norm": 0.5893413424491882, | |
| "learning_rate": 8.802029828000156e-05, | |
| "loss": 0.1718, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.04771134635455494, | |
| "grad_norm": 0.6724103093147278, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 0.1679, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.05009691367228269, | |
| "grad_norm": 0.8837454915046692, | |
| "learning_rate": 8.247240241650918e-05, | |
| "loss": 0.1861, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.05248248099001044, | |
| "grad_norm": 0.7627610564231873, | |
| "learning_rate": 7.938926261462366e-05, | |
| "loss": 0.1885, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.054868048307738185, | |
| "grad_norm": 0.47815459966659546, | |
| "learning_rate": 7.612492823579745e-05, | |
| "loss": 0.1022, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.05725361562546593, | |
| "grad_norm": 0.46670857071876526, | |
| "learning_rate": 7.269952498697734e-05, | |
| "loss": 0.1405, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.059639182943193675, | |
| "grad_norm": 0.5231714844703674, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 0.1264, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.06202475026092143, | |
| "grad_norm": 0.4806453585624695, | |
| "learning_rate": 6.545084971874738e-05, | |
| "loss": 0.1219, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.06202475026092143, | |
| "eval_loss": 0.0903618112206459, | |
| "eval_runtime": 16.6327, | |
| "eval_samples_per_second": 42.447, | |
| "eval_steps_per_second": 21.223, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.06441031757864918, | |
| "grad_norm": 0.386901319026947, | |
| "learning_rate": 6.167226819279528e-05, | |
| "loss": 0.1078, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.06679588489637692, | |
| "grad_norm": 0.4116002321243286, | |
| "learning_rate": 5.782172325201155e-05, | |
| "loss": 0.1013, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.06918145221410467, | |
| "grad_norm": 0.34669479727745056, | |
| "learning_rate": 5.392295478639225e-05, | |
| "loss": 0.1137, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.07156701953183241, | |
| "grad_norm": 0.3983478844165802, | |
| "learning_rate": 5e-05, | |
| "loss": 0.0615, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.07395258684956016, | |
| "grad_norm": 0.3107887804508209, | |
| "learning_rate": 4.607704521360776e-05, | |
| "loss": 0.0543, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.07633815416728791, | |
| "grad_norm": 0.3019241690635681, | |
| "learning_rate": 4.2178276747988446e-05, | |
| "loss": 0.0502, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.07872372148501565, | |
| "grad_norm": 0.33337318897247314, | |
| "learning_rate": 3.832773180720475e-05, | |
| "loss": 0.0531, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.0811092888027434, | |
| "grad_norm": 0.29022204875946045, | |
| "learning_rate": 3.4549150281252636e-05, | |
| "loss": 0.0534, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.08349485612047115, | |
| "grad_norm": 0.28613466024398804, | |
| "learning_rate": 3.086582838174551e-05, | |
| "loss": 0.0643, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.08588042343819889, | |
| "grad_norm": 0.29469338059425354, | |
| "learning_rate": 2.7300475013022663e-05, | |
| "loss": 0.0625, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.08826599075592664, | |
| "grad_norm": 0.31243982911109924, | |
| "learning_rate": 2.3875071764202563e-05, | |
| "loss": 0.053, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.0906515580736544, | |
| "grad_norm": 0.2379445880651474, | |
| "learning_rate": 2.061073738537635e-05, | |
| "loss": 0.0362, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.09303712539138213, | |
| "grad_norm": 0.21602892875671387, | |
| "learning_rate": 1.7527597583490822e-05, | |
| "loss": 0.0602, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.09303712539138213, | |
| "eval_loss": 0.05400439724326134, | |
| "eval_runtime": 16.603, | |
| "eval_samples_per_second": 42.522, | |
| "eval_steps_per_second": 21.261, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.09542269270910989, | |
| "grad_norm": 0.2588385045528412, | |
| "learning_rate": 1.4644660940672627e-05, | |
| "loss": 0.0422, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.09780826002683764, | |
| "grad_norm": 0.3525508940219879, | |
| "learning_rate": 1.1979701719998453e-05, | |
| "loss": 0.0671, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.10019382734456538, | |
| "grad_norm": 0.25098052620887756, | |
| "learning_rate": 9.549150281252633e-06, | |
| "loss": 0.0409, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.10257939466229313, | |
| "grad_norm": 0.3437732458114624, | |
| "learning_rate": 7.367991782295391e-06, | |
| "loss": 0.0647, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.10496496198002088, | |
| "grad_norm": 0.1902293562889099, | |
| "learning_rate": 5.449673790581611e-06, | |
| "loss": 0.0407, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.10735052929774862, | |
| "grad_norm": 0.2526785135269165, | |
| "learning_rate": 3.8060233744356633e-06, | |
| "loss": 0.0433, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.10973609661547637, | |
| "grad_norm": 0.33351102471351624, | |
| "learning_rate": 2.4471741852423237e-06, | |
| "loss": 0.0559, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.11212166393320412, | |
| "grad_norm": 0.23994620144367218, | |
| "learning_rate": 1.3815039801161721e-06, | |
| "loss": 0.0366, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.11450723125093186, | |
| "grad_norm": 0.23394052684307098, | |
| "learning_rate": 6.15582970243117e-07, | |
| "loss": 0.0506, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.11689279856865961, | |
| "grad_norm": 0.25432488322257996, | |
| "learning_rate": 1.5413331334360182e-07, | |
| "loss": 0.0285, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.11927836588638735, | |
| "grad_norm": 0.21557337045669556, | |
| "learning_rate": 0.0, | |
| "loss": 0.0358, | |
| "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": 3.2956458467328e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |