| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.7504690431519699, | |
| "eval_steps": 13, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0150093808630394, | |
| "grad_norm": 51.73713302612305, | |
| "learning_rate": 1e-05, | |
| "loss": 7.7969, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0150093808630394, | |
| "eval_loss": 7.751512050628662, | |
| "eval_runtime": 43.6829, | |
| "eval_samples_per_second": 10.279, | |
| "eval_steps_per_second": 1.305, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0300187617260788, | |
| "grad_norm": 48.8546028137207, | |
| "learning_rate": 2e-05, | |
| "loss": 7.8854, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.0450281425891182, | |
| "grad_norm": 43.629215240478516, | |
| "learning_rate": 3e-05, | |
| "loss": 7.4761, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0600375234521576, | |
| "grad_norm": 40.31449508666992, | |
| "learning_rate": 4e-05, | |
| "loss": 6.7166, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.075046904315197, | |
| "grad_norm": 31.518842697143555, | |
| "learning_rate": 5e-05, | |
| "loss": 6.0265, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.0900562851782364, | |
| "grad_norm": 27.299909591674805, | |
| "learning_rate": 6e-05, | |
| "loss": 5.1312, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.1050656660412758, | |
| "grad_norm": 24.453201293945312, | |
| "learning_rate": 7e-05, | |
| "loss": 4.2289, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.1200750469043152, | |
| "grad_norm": 21.977622985839844, | |
| "learning_rate": 8e-05, | |
| "loss": 3.409, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.1350844277673546, | |
| "grad_norm": 21.064271926879883, | |
| "learning_rate": 9e-05, | |
| "loss": 2.7192, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.150093808630394, | |
| "grad_norm": 18.23163414001465, | |
| "learning_rate": 0.0001, | |
| "loss": 1.9203, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.1651031894934334, | |
| "grad_norm": 18.346233367919922, | |
| "learning_rate": 9.98458666866564e-05, | |
| "loss": 1.3844, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.1801125703564728, | |
| "grad_norm": 17.36043357849121, | |
| "learning_rate": 9.938441702975689e-05, | |
| "loss": 1.0357, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.1951219512195122, | |
| "grad_norm": 6.1927690505981445, | |
| "learning_rate": 9.861849601988383e-05, | |
| "loss": 0.8596, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.1951219512195122, | |
| "eval_loss": 0.7876977324485779, | |
| "eval_runtime": 4.0203, | |
| "eval_samples_per_second": 111.684, | |
| "eval_steps_per_second": 14.178, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.2101313320825516, | |
| "grad_norm": 6.852001667022705, | |
| "learning_rate": 9.755282581475769e-05, | |
| "loss": 0.8059, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.225140712945591, | |
| "grad_norm": 15.55766487121582, | |
| "learning_rate": 9.619397662556435e-05, | |
| "loss": 0.7977, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.2401500938086304, | |
| "grad_norm": 6.66907262802124, | |
| "learning_rate": 9.45503262094184e-05, | |
| "loss": 0.7264, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.2551594746716698, | |
| "grad_norm": 6.188997745513916, | |
| "learning_rate": 9.263200821770461e-05, | |
| "loss": 0.7352, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.2701688555347092, | |
| "grad_norm": 2.3706018924713135, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 0.719, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.2851782363977486, | |
| "grad_norm": 5.199508190155029, | |
| "learning_rate": 8.802029828000156e-05, | |
| "loss": 0.755, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.300187617260788, | |
| "grad_norm": 2.7987897396087646, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 0.7206, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.3151969981238274, | |
| "grad_norm": 3.605015516281128, | |
| "learning_rate": 8.247240241650918e-05, | |
| "loss": 0.7075, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.3302063789868668, | |
| "grad_norm": 1.987611174583435, | |
| "learning_rate": 7.938926261462366e-05, | |
| "loss": 0.7034, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.3452157598499062, | |
| "grad_norm": 2.15010666847229, | |
| "learning_rate": 7.612492823579745e-05, | |
| "loss": 0.698, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.3602251407129456, | |
| "grad_norm": 2.248389959335327, | |
| "learning_rate": 7.269952498697734e-05, | |
| "loss": 0.717, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.37523452157598497, | |
| "grad_norm": 1.4804418087005615, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 0.707, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.3902439024390244, | |
| "grad_norm": 3.2395734786987305, | |
| "learning_rate": 6.545084971874738e-05, | |
| "loss": 0.708, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.3902439024390244, | |
| "eval_loss": 0.6942992806434631, | |
| "eval_runtime": 4.0219, | |
| "eval_samples_per_second": 111.638, | |
| "eval_steps_per_second": 14.172, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.4052532833020638, | |
| "grad_norm": 2.5209052562713623, | |
| "learning_rate": 6.167226819279528e-05, | |
| "loss": 0.7009, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.4202626641651032, | |
| "grad_norm": 2.8235585689544678, | |
| "learning_rate": 5.782172325201155e-05, | |
| "loss": 0.7067, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.4352720450281426, | |
| "grad_norm": 1.5787421464920044, | |
| "learning_rate": 5.392295478639225e-05, | |
| "loss": 0.707, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.450281425891182, | |
| "grad_norm": 3.7540314197540283, | |
| "learning_rate": 5e-05, | |
| "loss": 0.715, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.4652908067542214, | |
| "grad_norm": 3.978583574295044, | |
| "learning_rate": 4.607704521360776e-05, | |
| "loss": 0.7071, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.4803001876172608, | |
| "grad_norm": 1.993086338043213, | |
| "learning_rate": 4.2178276747988446e-05, | |
| "loss": 0.6994, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.49530956848030017, | |
| "grad_norm": 3.182072639465332, | |
| "learning_rate": 3.832773180720475e-05, | |
| "loss": 0.7108, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.5103189493433395, | |
| "grad_norm": 1.1114157438278198, | |
| "learning_rate": 3.4549150281252636e-05, | |
| "loss": 0.6889, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.525328330206379, | |
| "grad_norm": 1.384256362915039, | |
| "learning_rate": 3.086582838174551e-05, | |
| "loss": 0.7044, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.5403377110694184, | |
| "grad_norm": 3.324063777923584, | |
| "learning_rate": 2.7300475013022663e-05, | |
| "loss": 0.7098, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.5553470919324578, | |
| "grad_norm": 2.3583667278289795, | |
| "learning_rate": 2.3875071764202563e-05, | |
| "loss": 0.695, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.5703564727954972, | |
| "grad_norm": 1.7996037006378174, | |
| "learning_rate": 2.061073738537635e-05, | |
| "loss": 0.688, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.5853658536585366, | |
| "grad_norm": 0.9545679092407227, | |
| "learning_rate": 1.7527597583490822e-05, | |
| "loss": 0.6888, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.5853658536585366, | |
| "eval_loss": 0.6947711110115051, | |
| "eval_runtime": 5.7325, | |
| "eval_samples_per_second": 78.325, | |
| "eval_steps_per_second": 9.943, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.600375234521576, | |
| "grad_norm": 1.6693542003631592, | |
| "learning_rate": 1.4644660940672627e-05, | |
| "loss": 0.6815, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.6153846153846154, | |
| "grad_norm": 2.4521377086639404, | |
| "learning_rate": 1.1979701719998453e-05, | |
| "loss": 0.7027, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.6303939962476548, | |
| "grad_norm": 1.4812160730361938, | |
| "learning_rate": 9.549150281252633e-06, | |
| "loss": 0.7015, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.6454033771106942, | |
| "grad_norm": 2.221867799758911, | |
| "learning_rate": 7.367991782295391e-06, | |
| "loss": 0.6957, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.6604127579737336, | |
| "grad_norm": 3.274115800857544, | |
| "learning_rate": 5.449673790581611e-06, | |
| "loss": 0.7147, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.6754221388367729, | |
| "grad_norm": 1.3704664707183838, | |
| "learning_rate": 3.8060233744356633e-06, | |
| "loss": 0.6968, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.6904315196998124, | |
| "grad_norm": 1.5144388675689697, | |
| "learning_rate": 2.4471741852423237e-06, | |
| "loss": 0.7031, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.7054409005628518, | |
| "grad_norm": 2.387979030609131, | |
| "learning_rate": 1.3815039801161721e-06, | |
| "loss": 0.699, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.7204502814258912, | |
| "grad_norm": 1.8138819932937622, | |
| "learning_rate": 6.15582970243117e-07, | |
| "loss": 0.7033, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.7354596622889306, | |
| "grad_norm": 1.4316879510879517, | |
| "learning_rate": 1.5413331334360182e-07, | |
| "loss": 0.6938, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.7504690431519699, | |
| "grad_norm": 2.234066963195801, | |
| "learning_rate": 0.0, | |
| "loss": 0.7067, | |
| "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 | |
| } | |