| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.7027027027027026, | |
| "eval_steps": 17, | |
| "global_step": 200, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.013513513513513514, | |
| "eval_loss": 1.4550755023956299, | |
| "eval_runtime": 16.9459, | |
| "eval_samples_per_second": 7.376, | |
| "eval_steps_per_second": 0.944, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.04054054054054054, | |
| "grad_norm": 0.2170189917087555, | |
| "learning_rate": 3e-05, | |
| "loss": 1.4534, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.08108108108108109, | |
| "grad_norm": 0.25585711002349854, | |
| "learning_rate": 6e-05, | |
| "loss": 1.463, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.12162162162162163, | |
| "grad_norm": 0.24171331524848938, | |
| "learning_rate": 9e-05, | |
| "loss": 1.4157, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.16216216216216217, | |
| "grad_norm": 0.2852655351161957, | |
| "learning_rate": 9.997266286704631e-05, | |
| "loss": 1.3769, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.20270270270270271, | |
| "grad_norm": 0.3118453323841095, | |
| "learning_rate": 9.98292246503335e-05, | |
| "loss": 1.3167, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.22972972972972974, | |
| "eval_loss": 1.230002760887146, | |
| "eval_runtime": 17.2352, | |
| "eval_samples_per_second": 7.253, | |
| "eval_steps_per_second": 0.928, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.24324324324324326, | |
| "grad_norm": 0.3110101521015167, | |
| "learning_rate": 9.956320346634876e-05, | |
| "loss": 1.2565, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.28378378378378377, | |
| "grad_norm": 0.2778882086277008, | |
| "learning_rate": 9.917525374361912e-05, | |
| "loss": 1.216, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.32432432432432434, | |
| "grad_norm": 0.27365806698799133, | |
| "learning_rate": 9.86663298624003e-05, | |
| "loss": 1.2426, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.36486486486486486, | |
| "grad_norm": 0.27578839659690857, | |
| "learning_rate": 9.803768380684242e-05, | |
| "loss": 1.1636, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.40540540540540543, | |
| "grad_norm": 0.3146299123764038, | |
| "learning_rate": 9.729086208503174e-05, | |
| "loss": 1.1645, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.44594594594594594, | |
| "grad_norm": 0.3067275583744049, | |
| "learning_rate": 9.642770192448536e-05, | |
| "loss": 1.1556, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.4594594594594595, | |
| "eval_loss": 1.0813400745391846, | |
| "eval_runtime": 17.2384, | |
| "eval_samples_per_second": 7.251, | |
| "eval_steps_per_second": 0.928, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.4864864864864865, | |
| "grad_norm": 0.30459922552108765, | |
| "learning_rate": 9.545032675245813e-05, | |
| "loss": 1.1033, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.527027027027027, | |
| "grad_norm": 0.31739258766174316, | |
| "learning_rate": 9.43611409721806e-05, | |
| "loss": 1.0876, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.5675675675675675, | |
| "grad_norm": 0.3213581144809723, | |
| "learning_rate": 9.316282404787871e-05, | |
| "loss": 1.0733, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.6081081081081081, | |
| "grad_norm": 0.3462018370628357, | |
| "learning_rate": 9.185832391312644e-05, | |
| "loss": 1.1107, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.6486486486486487, | |
| "grad_norm": 0.3432579040527344, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 1.0265, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.6891891891891891, | |
| "grad_norm": 0.3420480191707611, | |
| "learning_rate": 8.894386393810563e-05, | |
| "loss": 1.053, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.6891891891891891, | |
| "eval_loss": 1.0123796463012695, | |
| "eval_runtime": 17.2325, | |
| "eval_samples_per_second": 7.254, | |
| "eval_steps_per_second": 0.928, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.7297297297297297, | |
| "grad_norm": 0.35601529479026794, | |
| "learning_rate": 8.73410738492077e-05, | |
| "loss": 1.0257, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.7702702702702703, | |
| "grad_norm": 0.3662780821323395, | |
| "learning_rate": 8.564642241456986e-05, | |
| "loss": 1.0528, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.8108108108108109, | |
| "grad_norm": 0.39324405789375305, | |
| "learning_rate": 8.386407858128706e-05, | |
| "loss": 1.0185, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.8513513513513513, | |
| "grad_norm": 0.39999616146087646, | |
| "learning_rate": 8.199842702516583e-05, | |
| "loss": 1.0132, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.8918918918918919, | |
| "grad_norm": 0.4472688138484955, | |
| "learning_rate": 8.005405736415126e-05, | |
| "loss": 1.0243, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.918918918918919, | |
| "eval_loss": 0.9729012250900269, | |
| "eval_runtime": 17.2325, | |
| "eval_samples_per_second": 7.254, | |
| "eval_steps_per_second": 0.928, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 0.9324324324324325, | |
| "grad_norm": 0.3893688917160034, | |
| "learning_rate": 7.803575286758364e-05, | |
| "loss": 0.9998, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.972972972972973, | |
| "grad_norm": 0.38515031337738037, | |
| "learning_rate": 7.594847868906076e-05, | |
| "loss": 0.9943, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 1.0135135135135136, | |
| "grad_norm": 0.4097621738910675, | |
| "learning_rate": 7.379736965185368e-05, | |
| "loss": 0.9972, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 1.054054054054054, | |
| "grad_norm": 0.4480155408382416, | |
| "learning_rate": 7.158771761692464e-05, | |
| "loss": 0.9534, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 1.0945945945945945, | |
| "grad_norm": 0.42299342155456543, | |
| "learning_rate": 6.932495846462261e-05, | |
| "loss": 0.9408, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 1.135135135135135, | |
| "grad_norm": 0.41073334217071533, | |
| "learning_rate": 6.701465872208216e-05, | |
| "loss": 0.9503, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 1.1486486486486487, | |
| "eval_loss": 0.9501034021377563, | |
| "eval_runtime": 17.2388, | |
| "eval_samples_per_second": 7.251, | |
| "eval_steps_per_second": 0.928, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 1.1756756756756757, | |
| "grad_norm": 0.49050870537757874, | |
| "learning_rate": 6.466250186922325e-05, | |
| "loss": 0.9146, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 1.2162162162162162, | |
| "grad_norm": 0.48001334071159363, | |
| "learning_rate": 6.227427435703997e-05, | |
| "loss": 0.9215, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.2567567567567568, | |
| "grad_norm": 0.47052550315856934, | |
| "learning_rate": 5.985585137257401e-05, | |
| "loss": 0.8942, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 1.2972972972972974, | |
| "grad_norm": 0.46515583992004395, | |
| "learning_rate": 5.74131823855921e-05, | |
| "loss": 0.9304, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 1.3378378378378377, | |
| "grad_norm": 0.5048130750656128, | |
| "learning_rate": 5.495227651252315e-05, | |
| "loss": 0.95, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 1.3783783783783785, | |
| "grad_norm": 0.5172950029373169, | |
| "learning_rate": 5.247918773366112e-05, | |
| "loss": 0.9288, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 1.3783783783783785, | |
| "eval_loss": 0.9384378790855408, | |
| "eval_runtime": 17.2324, | |
| "eval_samples_per_second": 7.254, | |
| "eval_steps_per_second": 0.928, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 1.4189189189189189, | |
| "grad_norm": 0.5409041047096252, | |
| "learning_rate": 5e-05, | |
| "loss": 0.9294, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 1.4594594594594594, | |
| "grad_norm": 0.49371689558029175, | |
| "learning_rate": 4.7520812266338885e-05, | |
| "loss": 0.9109, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "grad_norm": 0.4969201982021332, | |
| "learning_rate": 4.504772348747687e-05, | |
| "loss": 0.94, | |
| "step": 111 | |
| }, | |
| { | |
| "epoch": 1.5405405405405406, | |
| "grad_norm": 0.5152857899665833, | |
| "learning_rate": 4.2586817614407895e-05, | |
| "loss": 0.9083, | |
| "step": 114 | |
| }, | |
| { | |
| "epoch": 1.5810810810810811, | |
| "grad_norm": 0.5423870086669922, | |
| "learning_rate": 4.0144148627425993e-05, | |
| "loss": 0.8869, | |
| "step": 117 | |
| }, | |
| { | |
| "epoch": 1.6081081081081081, | |
| "eval_loss": 0.924616813659668, | |
| "eval_runtime": 17.244, | |
| "eval_samples_per_second": 7.249, | |
| "eval_steps_per_second": 0.928, | |
| "step": 119 | |
| }, | |
| { | |
| "epoch": 1.6216216216216215, | |
| "grad_norm": 0.49147549271583557, | |
| "learning_rate": 3.772572564296005e-05, | |
| "loss": 0.8855, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.6621621621621623, | |
| "grad_norm": 0.4987981915473938, | |
| "learning_rate": 3.533749813077677e-05, | |
| "loss": 0.8915, | |
| "step": 123 | |
| }, | |
| { | |
| "epoch": 1.7027027027027026, | |
| "grad_norm": 0.4984021484851837, | |
| "learning_rate": 3.298534127791785e-05, | |
| "loss": 0.9181, | |
| "step": 126 | |
| }, | |
| { | |
| "epoch": 1.7432432432432432, | |
| "grad_norm": 0.5324372053146362, | |
| "learning_rate": 3.0675041535377405e-05, | |
| "loss": 0.8911, | |
| "step": 129 | |
| }, | |
| { | |
| "epoch": 1.7837837837837838, | |
| "grad_norm": 0.521675169467926, | |
| "learning_rate": 2.8412282383075363e-05, | |
| "loss": 0.8905, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 1.8243243243243243, | |
| "grad_norm": 0.5264241099357605, | |
| "learning_rate": 2.6202630348146324e-05, | |
| "loss": 0.8928, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 1.8378378378378377, | |
| "eval_loss": 0.9162411689758301, | |
| "eval_runtime": 17.2393, | |
| "eval_samples_per_second": 7.251, | |
| "eval_steps_per_second": 0.928, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 1.864864864864865, | |
| "grad_norm": 0.5409694910049438, | |
| "learning_rate": 2.405152131093926e-05, | |
| "loss": 0.9095, | |
| "step": 138 | |
| }, | |
| { | |
| "epoch": 1.9054054054054053, | |
| "grad_norm": 0.5272465348243713, | |
| "learning_rate": 2.196424713241637e-05, | |
| "loss": 0.8844, | |
| "step": 141 | |
| }, | |
| { | |
| "epoch": 1.945945945945946, | |
| "grad_norm": 0.5436295866966248, | |
| "learning_rate": 1.9945942635848748e-05, | |
| "loss": 0.8699, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 1.9864864864864864, | |
| "grad_norm": 0.523435115814209, | |
| "learning_rate": 1.800157297483417e-05, | |
| "loss": 0.9164, | |
| "step": 147 | |
| }, | |
| { | |
| "epoch": 2.027027027027027, | |
| "grad_norm": 0.5417100787162781, | |
| "learning_rate": 1.6135921418712956e-05, | |
| "loss": 0.8899, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 2.0675675675675675, | |
| "grad_norm": 0.569843053817749, | |
| "learning_rate": 1.435357758543015e-05, | |
| "loss": 0.8476, | |
| "step": 153 | |
| }, | |
| { | |
| "epoch": 2.0675675675675675, | |
| "eval_loss": 0.9106208682060242, | |
| "eval_runtime": 17.255, | |
| "eval_samples_per_second": 7.244, | |
| "eval_steps_per_second": 0.927, | |
| "step": 153 | |
| }, | |
| { | |
| "epoch": 2.108108108108108, | |
| "grad_norm": 0.575161337852478, | |
| "learning_rate": 1.2658926150792322e-05, | |
| "loss": 0.8375, | |
| "step": 156 | |
| }, | |
| { | |
| "epoch": 2.1486486486486487, | |
| "grad_norm": 0.5137052536010742, | |
| "learning_rate": 1.1056136061894384e-05, | |
| "loss": 0.823, | |
| "step": 159 | |
| }, | |
| { | |
| "epoch": 2.189189189189189, | |
| "grad_norm": 0.532635509967804, | |
| "learning_rate": 9.549150281252633e-06, | |
| "loss": 0.8504, | |
| "step": 162 | |
| }, | |
| { | |
| "epoch": 2.22972972972973, | |
| "grad_norm": 0.5798172950744629, | |
| "learning_rate": 8.141676086873572e-06, | |
| "loss": 0.8315, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 2.27027027027027, | |
| "grad_norm": 0.5635726451873779, | |
| "learning_rate": 6.837175952121306e-06, | |
| "loss": 0.8626, | |
| "step": 168 | |
| }, | |
| { | |
| "epoch": 2.2972972972972974, | |
| "eval_loss": 0.9093061089515686, | |
| "eval_runtime": 17.2439, | |
| "eval_samples_per_second": 7.249, | |
| "eval_steps_per_second": 0.928, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 2.310810810810811, | |
| "grad_norm": 0.5467652082443237, | |
| "learning_rate": 5.6388590278194096e-06, | |
| "loss": 0.8334, | |
| "step": 171 | |
| }, | |
| { | |
| "epoch": 2.3513513513513513, | |
| "grad_norm": 0.534943699836731, | |
| "learning_rate": 4.549673247541875e-06, | |
| "loss": 0.8773, | |
| "step": 174 | |
| }, | |
| { | |
| "epoch": 2.391891891891892, | |
| "grad_norm": 0.5133833885192871, | |
| "learning_rate": 3.5722980755146517e-06, | |
| "loss": 0.8081, | |
| "step": 177 | |
| }, | |
| { | |
| "epoch": 2.4324324324324325, | |
| "grad_norm": 0.5191574096679688, | |
| "learning_rate": 2.7091379149682685e-06, | |
| "loss": 0.8595, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 2.472972972972973, | |
| "grad_norm": 0.5596672892570496, | |
| "learning_rate": 1.962316193157593e-06, | |
| "loss": 0.8525, | |
| "step": 183 | |
| }, | |
| { | |
| "epoch": 2.5135135135135136, | |
| "grad_norm": 0.5567631125450134, | |
| "learning_rate": 1.333670137599713e-06, | |
| "loss": 0.8269, | |
| "step": 186 | |
| }, | |
| { | |
| "epoch": 2.527027027027027, | |
| "eval_loss": 0.9086253643035889, | |
| "eval_runtime": 17.2348, | |
| "eval_samples_per_second": 7.253, | |
| "eval_steps_per_second": 0.928, | |
| "step": 187 | |
| }, | |
| { | |
| "epoch": 2.554054054054054, | |
| "grad_norm": 0.5464707016944885, | |
| "learning_rate": 8.247462563808817e-07, | |
| "loss": 0.8524, | |
| "step": 189 | |
| }, | |
| { | |
| "epoch": 2.5945945945945947, | |
| "grad_norm": 0.6008809804916382, | |
| "learning_rate": 4.367965336512403e-07, | |
| "loss": 0.8633, | |
| "step": 192 | |
| }, | |
| { | |
| "epoch": 2.635135135135135, | |
| "grad_norm": 0.5401892066001892, | |
| "learning_rate": 1.7077534966650766e-07, | |
| "loss": 0.8186, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 2.6756756756756754, | |
| "grad_norm": 0.5296628475189209, | |
| "learning_rate": 2.7337132953697554e-08, | |
| "loss": 0.8183, | |
| "step": 198 | |
| } | |
| ], | |
| "logging_steps": 3, | |
| "max_steps": 200, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 17, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 5.262308567427318e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |