| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.896, | |
| "eval_steps": 500, | |
| "global_step": 60, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.048, | |
| "grad_norm": 6.649672031402588, | |
| "learning_rate": 1.6666666666666667e-06, | |
| "loss": 1.2499, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.096, | |
| "grad_norm": 6.89929723739624, | |
| "learning_rate": 3.3333333333333333e-06, | |
| "loss": 1.2614, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.144, | |
| "grad_norm": 6.728121757507324, | |
| "learning_rate": 5e-06, | |
| "loss": 1.2755, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.192, | |
| "grad_norm": 4.547381401062012, | |
| "learning_rate": 6.666666666666667e-06, | |
| "loss": 1.167, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "grad_norm": 2.9474120140075684, | |
| "learning_rate": 8.333333333333334e-06, | |
| "loss": 1.1439, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.288, | |
| "grad_norm": 4.3956756591796875, | |
| "learning_rate": 1e-05, | |
| "loss": 1.164, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.336, | |
| "grad_norm": 4.672914505004883, | |
| "learning_rate": 9.991540791356342e-06, | |
| "loss": 1.193, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.384, | |
| "grad_norm": 3.9675629138946533, | |
| "learning_rate": 9.966191788709716e-06, | |
| "loss": 1.0955, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.432, | |
| "grad_norm": 4.041304111480713, | |
| "learning_rate": 9.924038765061042e-06, | |
| "loss": 1.138, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 2.6356163024902344, | |
| "learning_rate": 9.86522435289912e-06, | |
| "loss": 1.1076, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.528, | |
| "grad_norm": 1.9111108779907227, | |
| "learning_rate": 9.789947561577445e-06, | |
| "loss": 1.0336, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.576, | |
| "grad_norm": 1.8051618337631226, | |
| "learning_rate": 9.698463103929542e-06, | |
| "loss": 1.0237, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.624, | |
| "grad_norm": 1.7929407358169556, | |
| "learning_rate": 9.591080534401371e-06, | |
| "loss": 1.0179, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.672, | |
| "grad_norm": 1.2241275310516357, | |
| "learning_rate": 9.468163201617063e-06, | |
| "loss": 1.0027, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "grad_norm": 1.1858434677124023, | |
| "learning_rate": 9.330127018922195e-06, | |
| "loss": 0.9925, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.768, | |
| "grad_norm": 1.1306203603744507, | |
| "learning_rate": 9.177439057064684e-06, | |
| "loss": 0.9878, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.816, | |
| "grad_norm": 0.8818135857582092, | |
| "learning_rate": 9.01061596377522e-06, | |
| "loss": 0.9923, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.864, | |
| "grad_norm": 0.9130216836929321, | |
| "learning_rate": 8.83022221559489e-06, | |
| "loss": 0.9749, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.912, | |
| "grad_norm": 0.9632487297058105, | |
| "learning_rate": 8.636868207865244e-06, | |
| "loss": 0.9665, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 1.004268765449524, | |
| "learning_rate": 8.43120818934367e-06, | |
| "loss": 0.9803, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.016, | |
| "grad_norm": 1.14327871799469, | |
| "learning_rate": 8.213938048432697e-06, | |
| "loss": 1.2317, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 1.064, | |
| "grad_norm": 0.8447701930999756, | |
| "learning_rate": 7.985792958513932e-06, | |
| "loss": 0.8934, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 1.112, | |
| "grad_norm": 0.9320511221885681, | |
| "learning_rate": 7.747544890354031e-06, | |
| "loss": 0.991, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "grad_norm": 0.8384785652160645, | |
| "learning_rate": 7.500000000000001e-06, | |
| "loss": 0.9389, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 1.208, | |
| "grad_norm": 0.6830297112464905, | |
| "learning_rate": 7.243995901002312e-06, | |
| "loss": 0.86, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 1.256, | |
| "grad_norm": 0.7668797373771667, | |
| "learning_rate": 6.980398830195785e-06, | |
| "loss": 0.9331, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 1.304, | |
| "grad_norm": 0.805425226688385, | |
| "learning_rate": 6.710100716628345e-06, | |
| "loss": 0.9881, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 1.3519999999999999, | |
| "grad_norm": 0.6650639772415161, | |
| "learning_rate": 6.434016163555452e-06, | |
| "loss": 0.9337, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "grad_norm": 0.6443905234336853, | |
| "learning_rate": 6.153079353712201e-06, | |
| "loss": 0.8651, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 1.448, | |
| "grad_norm": 0.7086542844772339, | |
| "learning_rate": 5.8682408883346535e-06, | |
| "loss": 0.8596, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.496, | |
| "grad_norm": 0.5941119194030762, | |
| "learning_rate": 5.5804645706261515e-06, | |
| "loss": 0.8963, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 1.544, | |
| "grad_norm": 0.6889335513114929, | |
| "learning_rate": 5.290724144552379e-06, | |
| "loss": 0.8475, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 1.592, | |
| "grad_norm": 0.6347829699516296, | |
| "learning_rate": 5e-06, | |
| "loss": 0.8904, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 1.6400000000000001, | |
| "grad_norm": 0.681000828742981, | |
| "learning_rate": 4.7092758554476215e-06, | |
| "loss": 0.9363, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 1.688, | |
| "grad_norm": 0.604396641254425, | |
| "learning_rate": 4.4195354293738484e-06, | |
| "loss": 0.9859, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 1.736, | |
| "grad_norm": 0.5908040404319763, | |
| "learning_rate": 4.131759111665349e-06, | |
| "loss": 0.7657, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 1.784, | |
| "grad_norm": 0.6690030694007874, | |
| "learning_rate": 3.8469206462878e-06, | |
| "loss": 0.9692, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 1.8319999999999999, | |
| "grad_norm": 0.5841237306594849, | |
| "learning_rate": 3.5659838364445505e-06, | |
| "loss": 0.894, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "grad_norm": 0.6076769828796387, | |
| "learning_rate": 3.289899283371657e-06, | |
| "loss": 1.0476, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 1.928, | |
| "grad_norm": 0.54935222864151, | |
| "learning_rate": 3.019601169804216e-06, | |
| "loss": 0.7547, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.976, | |
| "grad_norm": 0.7094786763191223, | |
| "learning_rate": 2.7560040989976894e-06, | |
| "loss": 1.0386, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 2.032, | |
| "grad_norm": 0.7260507941246033, | |
| "learning_rate": 2.5000000000000015e-06, | |
| "loss": 1.0944, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "grad_norm": 0.5274712443351746, | |
| "learning_rate": 2.2524551096459703e-06, | |
| "loss": 0.7287, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 2.128, | |
| "grad_norm": 0.5840578675270081, | |
| "learning_rate": 2.0142070414860704e-06, | |
| "loss": 0.9188, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 2.176, | |
| "grad_norm": 0.5778673887252808, | |
| "learning_rate": 1.7860619515673034e-06, | |
| "loss": 0.8059, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 2.224, | |
| "grad_norm": 0.541653037071228, | |
| "learning_rate": 1.5687918106563326e-06, | |
| "loss": 0.9784, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 2.2720000000000002, | |
| "grad_norm": 0.4745509922504425, | |
| "learning_rate": 1.3631317921347564e-06, | |
| "loss": 0.832, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 2.32, | |
| "grad_norm": 0.5039135813713074, | |
| "learning_rate": 1.1697777844051105e-06, | |
| "loss": 0.8944, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 2.368, | |
| "grad_norm": 0.5412746071815491, | |
| "learning_rate": 9.893840362247809e-07, | |
| "loss": 0.9197, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 2.416, | |
| "grad_norm": 0.481842964887619, | |
| "learning_rate": 8.225609429353187e-07, | |
| "loss": 0.8418, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 2.464, | |
| "grad_norm": 0.4687838554382324, | |
| "learning_rate": 6.698729810778065e-07, | |
| "loss": 0.8105, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 2.512, | |
| "grad_norm": 0.5128464698791504, | |
| "learning_rate": 5.318367983829393e-07, | |
| "loss": 0.8783, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 2.56, | |
| "grad_norm": 0.48326826095581055, | |
| "learning_rate": 4.089194655986306e-07, | |
| "loss": 0.8691, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 2.608, | |
| "grad_norm": 0.531957745552063, | |
| "learning_rate": 3.015368960704584e-07, | |
| "loss": 0.8631, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 2.656, | |
| "grad_norm": 0.4845496714115143, | |
| "learning_rate": 2.1005243842255552e-07, | |
| "loss": 0.8204, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 2.7039999999999997, | |
| "grad_norm": 0.47783514857292175, | |
| "learning_rate": 1.3477564710088097e-07, | |
| "loss": 0.8359, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 2.752, | |
| "grad_norm": 0.5208674073219299, | |
| "learning_rate": 7.59612349389599e-08, | |
| "loss": 0.9702, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "grad_norm": 0.4236156940460205, | |
| "learning_rate": 3.3808211290284886e-08, | |
| "loss": 0.7322, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 2.848, | |
| "grad_norm": 0.47887781262397766, | |
| "learning_rate": 8.459208643659122e-09, | |
| "loss": 0.9654, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 2.896, | |
| "grad_norm": 0.43384796380996704, | |
| "learning_rate": 0.0, | |
| "loss": 0.8416, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 2.896, | |
| "step": 60, | |
| "total_flos": 59731024576512.0, | |
| "train_loss": 0.9648277342319489, | |
| "train_runtime": 3605.7128, | |
| "train_samples_per_second": 1.664, | |
| "train_steps_per_second": 0.017 | |
| } | |
| ], | |
| "logging_steps": 1.0, | |
| "max_steps": 60, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 59731024576512.0, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |