| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.96, | |
| "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": 0.5960447192192078, | |
| "learning_rate": 1.6666666666666667e-06, | |
| "loss": 0.6433, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.096, | |
| "grad_norm": 0.5785825252532959, | |
| "learning_rate": 3.3333333333333333e-06, | |
| "loss": 0.637, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.144, | |
| "grad_norm": 0.6395981311798096, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6437, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.192, | |
| "grad_norm": 0.32435280084609985, | |
| "learning_rate": 6.666666666666667e-06, | |
| "loss": 0.5838, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "grad_norm": 0.3350636661052704, | |
| "learning_rate": 8.333333333333334e-06, | |
| "loss": 0.5055, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.288, | |
| "grad_norm": 0.26381659507751465, | |
| "learning_rate": 1e-05, | |
| "loss": 0.4363, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.336, | |
| "grad_norm": 0.2484428882598877, | |
| "learning_rate": 9.991540791356342e-06, | |
| "loss": 0.4857, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.384, | |
| "grad_norm": 0.2681486904621124, | |
| "learning_rate": 9.966191788709716e-06, | |
| "loss": 0.5556, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.432, | |
| "grad_norm": 0.28722772002220154, | |
| "learning_rate": 9.924038765061042e-06, | |
| "loss": 0.5532, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 0.26680728793144226, | |
| "learning_rate": 9.86522435289912e-06, | |
| "loss": 0.6138, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.528, | |
| "grad_norm": 0.21414291858673096, | |
| "learning_rate": 9.789947561577445e-06, | |
| "loss": 0.5207, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.576, | |
| "grad_norm": 0.22129523754119873, | |
| "learning_rate": 9.698463103929542e-06, | |
| "loss": 0.5395, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.624, | |
| "grad_norm": 0.2071954756975174, | |
| "learning_rate": 9.591080534401371e-06, | |
| "loss": 0.5599, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.672, | |
| "grad_norm": 0.18173572421073914, | |
| "learning_rate": 9.468163201617063e-06, | |
| "loss": 0.4935, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "grad_norm": 0.1673681139945984, | |
| "learning_rate": 9.330127018922195e-06, | |
| "loss": 0.5062, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.768, | |
| "grad_norm": 0.1739712953567505, | |
| "learning_rate": 9.177439057064684e-06, | |
| "loss": 0.5003, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.816, | |
| "grad_norm": 0.16390115022659302, | |
| "learning_rate": 9.01061596377522e-06, | |
| "loss": 0.4908, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.864, | |
| "grad_norm": 0.1472504436969757, | |
| "learning_rate": 8.83022221559489e-06, | |
| "loss": 0.4909, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.912, | |
| "grad_norm": 0.14271003007888794, | |
| "learning_rate": 8.636868207865244e-06, | |
| "loss": 0.4884, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 0.15660056471824646, | |
| "learning_rate": 8.43120818934367e-06, | |
| "loss": 0.4936, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.048, | |
| "grad_norm": 0.2163340151309967, | |
| "learning_rate": 8.213938048432697e-06, | |
| "loss": 0.7499, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 1.096, | |
| "grad_norm": 0.14213234186172485, | |
| "learning_rate": 7.985792958513932e-06, | |
| "loss": 0.3572, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 1.144, | |
| "grad_norm": 0.14226825535297394, | |
| "learning_rate": 7.747544890354031e-06, | |
| "loss": 0.3631, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 1.192, | |
| "grad_norm": 0.12011696398258209, | |
| "learning_rate": 7.500000000000001e-06, | |
| "loss": 0.3414, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "grad_norm": 0.14066629111766815, | |
| "learning_rate": 7.243995901002312e-06, | |
| "loss": 0.3472, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 1.288, | |
| "grad_norm": 0.11682265251874924, | |
| "learning_rate": 6.980398830195785e-06, | |
| "loss": 0.3415, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 1.336, | |
| "grad_norm": 0.12975744903087616, | |
| "learning_rate": 6.710100716628345e-06, | |
| "loss": 0.3328, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 1.384, | |
| "grad_norm": 0.11025315523147583, | |
| "learning_rate": 6.434016163555452e-06, | |
| "loss": 0.3587, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 1.432, | |
| "grad_norm": 0.12963436543941498, | |
| "learning_rate": 6.153079353712201e-06, | |
| "loss": 0.3098, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 1.48, | |
| "grad_norm": 0.11641515046358109, | |
| "learning_rate": 5.8682408883346535e-06, | |
| "loss": 0.3159, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.528, | |
| "grad_norm": 0.1281406134366989, | |
| "learning_rate": 5.5804645706261515e-06, | |
| "loss": 0.3283, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 1.576, | |
| "grad_norm": 0.11018883436918259, | |
| "learning_rate": 5.290724144552379e-06, | |
| "loss": 0.3175, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 1.624, | |
| "grad_norm": 0.13282187283039093, | |
| "learning_rate": 5e-06, | |
| "loss": 0.2702, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 1.6720000000000002, | |
| "grad_norm": 0.14893370866775513, | |
| "learning_rate": 4.7092758554476215e-06, | |
| "loss": 0.2807, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 1.72, | |
| "grad_norm": 0.1253272294998169, | |
| "learning_rate": 4.4195354293738484e-06, | |
| "loss": 0.2936, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 1.768, | |
| "grad_norm": 0.11152789741754532, | |
| "learning_rate": 4.131759111665349e-06, | |
| "loss": 0.3005, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 1.8159999999999998, | |
| "grad_norm": 0.11078399419784546, | |
| "learning_rate": 3.8469206462878e-06, | |
| "loss": 0.3255, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 1.8639999999999999, | |
| "grad_norm": 0.11652141064405441, | |
| "learning_rate": 3.5659838364445505e-06, | |
| "loss": 0.3027, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 1.912, | |
| "grad_norm": 0.10897649079561234, | |
| "learning_rate": 3.289899283371657e-06, | |
| "loss": 0.2728, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 1.96, | |
| "grad_norm": 0.1067444458603859, | |
| "learning_rate": 3.019601169804216e-06, | |
| "loss": 0.3053, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 2.048, | |
| "grad_norm": 0.12603534758090973, | |
| "learning_rate": 2.7560040989976894e-06, | |
| "loss": 0.5317, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 2.096, | |
| "grad_norm": 0.10293126851320267, | |
| "learning_rate": 2.5000000000000015e-06, | |
| "loss": 0.2568, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 2.144, | |
| "grad_norm": 0.09188748151063919, | |
| "learning_rate": 2.2524551096459703e-06, | |
| "loss": 0.208, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 2.192, | |
| "grad_norm": 0.08503475040197372, | |
| "learning_rate": 2.0142070414860704e-06, | |
| "loss": 0.2053, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 2.24, | |
| "grad_norm": 0.10515239089727402, | |
| "learning_rate": 1.7860619515673034e-06, | |
| "loss": 0.2588, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 2.288, | |
| "grad_norm": 0.08518028259277344, | |
| "learning_rate": 1.5687918106563326e-06, | |
| "loss": 0.2036, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 2.336, | |
| "grad_norm": 0.08311820775270462, | |
| "learning_rate": 1.3631317921347564e-06, | |
| "loss": 0.2469, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 2.384, | |
| "grad_norm": 0.08776900172233582, | |
| "learning_rate": 1.1697777844051105e-06, | |
| "loss": 0.2237, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 2.432, | |
| "grad_norm": 0.09390392154455185, | |
| "learning_rate": 9.893840362247809e-07, | |
| "loss": 0.2476, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 2.48, | |
| "grad_norm": 0.09852010011672974, | |
| "learning_rate": 8.225609429353187e-07, | |
| "loss": 0.2475, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 2.528, | |
| "grad_norm": 0.10095316171646118, | |
| "learning_rate": 6.698729810778065e-07, | |
| "loss": 0.2274, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 2.576, | |
| "grad_norm": 0.08666174858808517, | |
| "learning_rate": 5.318367983829393e-07, | |
| "loss": 0.2497, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 2.624, | |
| "grad_norm": 0.10396892577409744, | |
| "learning_rate": 4.089194655986306e-07, | |
| "loss": 0.2269, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 2.672, | |
| "grad_norm": 0.09909585118293762, | |
| "learning_rate": 3.015368960704584e-07, | |
| "loss": 0.2198, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 2.7199999999999998, | |
| "grad_norm": 0.08612281084060669, | |
| "learning_rate": 2.1005243842255552e-07, | |
| "loss": 0.2219, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 2.768, | |
| "grad_norm": 0.09808117151260376, | |
| "learning_rate": 1.3477564710088097e-07, | |
| "loss": 0.209, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 2.816, | |
| "grad_norm": 0.09213282912969589, | |
| "learning_rate": 7.59612349389599e-08, | |
| "loss": 0.2084, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 2.864, | |
| "grad_norm": 0.10377876460552216, | |
| "learning_rate": 3.3808211290284886e-08, | |
| "loss": 0.2487, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 2.912, | |
| "grad_norm": 0.10324692726135254, | |
| "learning_rate": 8.459208643659122e-09, | |
| "loss": 0.2267, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 2.96, | |
| "grad_norm": 0.09482261538505554, | |
| "learning_rate": 0.0, | |
| "loss": 0.2104, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 2.96, | |
| "step": 60, | |
| "total_flos": 2.770915149689651e+16, | |
| "train_loss": 0.37391861056288084, | |
| "train_runtime": 21254.7027, | |
| "train_samples_per_second": 0.071, | |
| "train_steps_per_second": 0.003 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 60, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 200, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2.770915149689651e+16, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |