| { | |
| "best_global_step": null, | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.9910313901345291, | |
| "eval_steps": 500, | |
| "global_step": 444, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.02242152466367713, | |
| "grad_norm": 2.15625, | |
| "learning_rate": 2.305093438544906e-05, | |
| "loss": 0.9112, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.04484304932735426, | |
| "grad_norm": 1.4765625, | |
| "learning_rate": 5.1864602367260384e-05, | |
| "loss": 0.7142, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.06726457399103139, | |
| "grad_norm": 1.09375, | |
| "learning_rate": 8.067827034907172e-05, | |
| "loss": 0.6696, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.08968609865470852, | |
| "grad_norm": 0.9921875, | |
| "learning_rate": 0.00010949193833088303, | |
| "loss": 0.652, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.11210762331838565, | |
| "grad_norm": 0.87109375, | |
| "learning_rate": 0.00013830560631269436, | |
| "loss": 0.6583, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.13452914798206278, | |
| "grad_norm": 1.2578125, | |
| "learning_rate": 0.00016711927429450567, | |
| "loss": 0.6354, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.15695067264573992, | |
| "grad_norm": 1.1875, | |
| "learning_rate": 0.000195932942276317, | |
| "loss": 0.6436, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.17937219730941703, | |
| "grad_norm": 1.1484375, | |
| "learning_rate": 0.00022474661025812833, | |
| "loss": 0.6563, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.20179372197309417, | |
| "grad_norm": 9.5, | |
| "learning_rate": 0.00025356027823993967, | |
| "loss": 0.8176, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.2242152466367713, | |
| "grad_norm": 1.2890625, | |
| "learning_rate": 0.00028237394622175095, | |
| "loss": 0.8045, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.24663677130044842, | |
| "grad_norm": 1.3125, | |
| "learning_rate": 0.00028811441478558746, | |
| "loss": 0.7267, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.26905829596412556, | |
| "grad_norm": 1.2578125, | |
| "learning_rate": 0.0002880239788169296, | |
| "loss": 0.7135, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.2914798206278027, | |
| "grad_norm": 0.6953125, | |
| "learning_rate": 0.0002878640385367635, | |
| "loss": 0.6942, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.31390134529147984, | |
| "grad_norm": 0.640625, | |
| "learning_rate": 0.0002876346969346014, | |
| "loss": 0.6893, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.336322869955157, | |
| "grad_norm": 0.66015625, | |
| "learning_rate": 0.0002873361016891883, | |
| "loss": 0.6921, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.35874439461883406, | |
| "grad_norm": 0.60546875, | |
| "learning_rate": 0.000286968445073407, | |
| "loss": 0.6859, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.3811659192825112, | |
| "grad_norm": 0.57421875, | |
| "learning_rate": 0.00028653196383046967, | |
| "loss": 0.6619, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 0.40358744394618834, | |
| "grad_norm": 0.54296875, | |
| "learning_rate": 0.0002860269390214723, | |
| "loss": 0.6673, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.4260089686098655, | |
| "grad_norm": 0.55859375, | |
| "learning_rate": 0.00028545369584441287, | |
| "loss": 0.6669, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 0.4484304932735426, | |
| "grad_norm": 0.6328125, | |
| "learning_rate": 0.00028481260342478823, | |
| "loss": 0.6705, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.47085201793721976, | |
| "grad_norm": 0.53515625, | |
| "learning_rate": 0.000284104074577905, | |
| "loss": 0.6551, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 0.49327354260089684, | |
| "grad_norm": 0.5390625, | |
| "learning_rate": 0.00028332856554305765, | |
| "loss": 0.6536, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.515695067264574, | |
| "grad_norm": 0.484375, | |
| "learning_rate": 0.0002824865756897446, | |
| "loss": 0.6496, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 0.5381165919282511, | |
| "grad_norm": 0.5, | |
| "learning_rate": 0.0002815786471961118, | |
| "loss": 0.6421, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.5605381165919282, | |
| "grad_norm": 0.44921875, | |
| "learning_rate": 0.00028060536469983084, | |
| "loss": 0.6333, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.5829596412556054, | |
| "grad_norm": 0.46484375, | |
| "learning_rate": 0.0002795673549216364, | |
| "loss": 0.6506, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.6053811659192825, | |
| "grad_norm": 0.515625, | |
| "learning_rate": 0.0002784652862617649, | |
| "loss": 0.6242, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 0.6278026905829597, | |
| "grad_norm": 0.5234375, | |
| "learning_rate": 0.0002772998683695552, | |
| "loss": 0.6467, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.6502242152466368, | |
| "grad_norm": 0.43359375, | |
| "learning_rate": 0.00027607185168648785, | |
| "loss": 0.634, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 0.672645739910314, | |
| "grad_norm": 0.5078125, | |
| "learning_rate": 0.0002747820269629572, | |
| "loss": 0.6367, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.695067264573991, | |
| "grad_norm": 0.478515625, | |
| "learning_rate": 0.0002734312247490874, | |
| "loss": 0.6091, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 0.7174887892376681, | |
| "grad_norm": 0.421875, | |
| "learning_rate": 0.0002720203148599208, | |
| "loss": 0.6293, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.7399103139013453, | |
| "grad_norm": 0.4140625, | |
| "learning_rate": 0.00027055020581532246, | |
| "loss": 0.6168, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 0.7623318385650224, | |
| "grad_norm": 0.462890625, | |
| "learning_rate": 0.00026902184425496155, | |
| "loss": 0.6284, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.7847533632286996, | |
| "grad_norm": 0.435546875, | |
| "learning_rate": 0.0002674362143287467, | |
| "loss": 0.6163, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.8071748878923767, | |
| "grad_norm": 0.47265625, | |
| "learning_rate": 0.0002657943370631075, | |
| "loss": 0.6099, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.8295964125560538, | |
| "grad_norm": 0.53125, | |
| "learning_rate": 0.00026409726970353, | |
| "loss": 0.6159, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 0.852017937219731, | |
| "grad_norm": 0.48828125, | |
| "learning_rate": 0.00026234610503377063, | |
| "loss": 0.6001, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.874439461883408, | |
| "grad_norm": 0.41796875, | |
| "learning_rate": 0.00026054197067218514, | |
| "loss": 0.5967, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 0.8968609865470852, | |
| "grad_norm": 0.435546875, | |
| "learning_rate": 0.0002586860283456274, | |
| "loss": 0.6301, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.9192825112107623, | |
| "grad_norm": 0.447265625, | |
| "learning_rate": 0.00025677947314138464, | |
| "loss": 0.6061, | |
| "step": 205 | |
| }, | |
| { | |
| "epoch": 0.9417040358744395, | |
| "grad_norm": 0.447265625, | |
| "learning_rate": 0.00025482353273763113, | |
| "loss": 0.5923, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.9641255605381166, | |
| "grad_norm": 0.494140625, | |
| "learning_rate": 0.0002528194666128958, | |
| "loss": 0.6004, | |
| "step": 215 | |
| }, | |
| { | |
| "epoch": 0.9865470852017937, | |
| "grad_norm": 0.419921875, | |
| "learning_rate": 0.0002507685652350527, | |
| "loss": 0.612, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.9955156950672646, | |
| "eval_loss": 0.6039933562278748, | |
| "eval_runtime": 2.2333, | |
| "eval_samples_per_second": 18.806, | |
| "eval_steps_per_second": 18.806, | |
| "step": 222 | |
| }, | |
| { | |
| "epoch": 1.0089686098654709, | |
| "grad_norm": 0.55859375, | |
| "learning_rate": 0.0002486721492303566, | |
| "loss": 0.533, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 1.031390134529148, | |
| "grad_norm": 0.451171875, | |
| "learning_rate": 0.0002465315685330595, | |
| "loss": 0.4381, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 1.053811659192825, | |
| "grad_norm": 0.466796875, | |
| "learning_rate": 0.0002443482015161539, | |
| "loss": 0.4257, | |
| "step": 235 | |
| }, | |
| { | |
| "epoch": 1.0762331838565022, | |
| "grad_norm": 0.490234375, | |
| "learning_rate": 0.0002421234541038045, | |
| "loss": 0.4324, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.0986547085201794, | |
| "grad_norm": 0.404296875, | |
| "learning_rate": 0.00023985875886603888, | |
| "loss": 0.4228, | |
| "step": 245 | |
| }, | |
| { | |
| "epoch": 1.1210762331838564, | |
| "grad_norm": 0.40234375, | |
| "learning_rate": 0.00023755557409627998, | |
| "loss": 0.4219, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.1434977578475336, | |
| "grad_norm": 0.373046875, | |
| "learning_rate": 0.00023521538287231476, | |
| "loss": 0.4365, | |
| "step": 255 | |
| }, | |
| { | |
| "epoch": 1.1659192825112108, | |
| "grad_norm": 0.42578125, | |
| "learning_rate": 0.0002328396921013038, | |
| "loss": 0.4315, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 1.188340807174888, | |
| "grad_norm": 0.384765625, | |
| "learning_rate": 0.00023043003154944643, | |
| "loss": 0.428, | |
| "step": 265 | |
| }, | |
| { | |
| "epoch": 1.210762331838565, | |
| "grad_norm": 0.41796875, | |
| "learning_rate": 0.000227987952856926, | |
| "loss": 0.4248, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 1.2331838565022422, | |
| "grad_norm": 0.365234375, | |
| "learning_rate": 0.00022551502853877082, | |
| "loss": 0.4306, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 1.2556053811659194, | |
| "grad_norm": 0.376953125, | |
| "learning_rate": 0.00022301285097227208, | |
| "loss": 0.4423, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 1.2780269058295963, | |
| "grad_norm": 0.39453125, | |
| "learning_rate": 0.00022048303137161342, | |
| "loss": 0.4351, | |
| "step": 285 | |
| }, | |
| { | |
| "epoch": 1.3004484304932735, | |
| "grad_norm": 0.3828125, | |
| "learning_rate": 0.00021792719875036935, | |
| "loss": 0.4371, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 1.3228699551569507, | |
| "grad_norm": 0.380859375, | |
| "learning_rate": 0.00021534699887254367, | |
| "loss": 0.434, | |
| "step": 295 | |
| }, | |
| { | |
| "epoch": 1.3452914798206277, | |
| "grad_norm": 0.375, | |
| "learning_rate": 0.00021274409319282082, | |
| "loss": 0.4292, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.3677130044843049, | |
| "grad_norm": 0.37890625, | |
| "learning_rate": 0.00021012015778671412, | |
| "loss": 0.4345, | |
| "step": 305 | |
| }, | |
| { | |
| "epoch": 1.390134529147982, | |
| "grad_norm": 0.390625, | |
| "learning_rate": 0.00020747688227129932, | |
| "loss": 0.4313, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 1.4125560538116593, | |
| "grad_norm": 0.3671875, | |
| "learning_rate": 0.00020481596871722803, | |
| "loss": 0.4319, | |
| "step": 315 | |
| }, | |
| { | |
| "epoch": 1.4349775784753362, | |
| "grad_norm": 0.42578125, | |
| "learning_rate": 0.0002021391305527223, | |
| "loss": 0.4306, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 1.4573991031390134, | |
| "grad_norm": 0.359375, | |
| "learning_rate": 0.00019944809146025586, | |
| "loss": 0.4415, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 1.4798206278026906, | |
| "grad_norm": 0.376953125, | |
| "learning_rate": 0.00019674458426663204, | |
| "loss": 0.4288, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 1.5022421524663678, | |
| "grad_norm": 0.365234375, | |
| "learning_rate": 0.0001940303498271737, | |
| "loss": 0.4263, | |
| "step": 335 | |
| }, | |
| { | |
| "epoch": 1.5246636771300448, | |
| "grad_norm": 0.361328125, | |
| "learning_rate": 0.00019130713590474326, | |
| "loss": 0.429, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 1.547085201793722, | |
| "grad_norm": 0.3671875, | |
| "learning_rate": 0.00018857669604431496, | |
| "loss": 0.4147, | |
| "step": 345 | |
| }, | |
| { | |
| "epoch": 1.5695067264573992, | |
| "grad_norm": 0.369140625, | |
| "learning_rate": 0.0001858407884438237, | |
| "loss": 0.4215, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.5919282511210762, | |
| "grad_norm": 0.373046875, | |
| "learning_rate": 0.0001831011748220177, | |
| "loss": 0.4232, | |
| "step": 355 | |
| }, | |
| { | |
| "epoch": 1.6143497757847534, | |
| "grad_norm": 0.359375, | |
| "learning_rate": 0.00018035961928404432, | |
| "loss": 0.4133, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 1.6367713004484306, | |
| "grad_norm": 0.3515625, | |
| "learning_rate": 0.0001776178871854989, | |
| "loss": 0.4198, | |
| "step": 365 | |
| }, | |
| { | |
| "epoch": 1.6591928251121075, | |
| "grad_norm": 0.35546875, | |
| "learning_rate": 0.00017487774399566828, | |
| "loss": 0.4253, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 1.6816143497757847, | |
| "grad_norm": 0.36328125, | |
| "learning_rate": 0.00017214095416070196, | |
| "loss": 0.4247, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 1.704035874439462, | |
| "grad_norm": 0.3515625, | |
| "learning_rate": 0.00016940927996744084, | |
| "loss": 0.4173, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 1.726457399103139, | |
| "grad_norm": 0.337890625, | |
| "learning_rate": 0.00016668448040863722, | |
| "loss": 0.4197, | |
| "step": 385 | |
| }, | |
| { | |
| "epoch": 1.7488789237668163, | |
| "grad_norm": 0.357421875, | |
| "learning_rate": 0.000163968310050296, | |
| "loss": 0.4171, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 1.7713004484304933, | |
| "grad_norm": 0.33203125, | |
| "learning_rate": 0.00016126251790186578, | |
| "loss": 0.3982, | |
| "step": 395 | |
| }, | |
| { | |
| "epoch": 1.7937219730941703, | |
| "grad_norm": 0.330078125, | |
| "learning_rate": 0.0001585688462900084, | |
| "loss": 0.4102, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.8161434977578477, | |
| "grad_norm": 0.341796875, | |
| "learning_rate": 0.0001558890297366718, | |
| "loss": 0.4096, | |
| "step": 405 | |
| }, | |
| { | |
| "epoch": 1.8385650224215246, | |
| "grad_norm": 0.330078125, | |
| "learning_rate": 0.000153224793842188, | |
| "loss": 0.4118, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 1.8609865470852018, | |
| "grad_norm": 0.328125, | |
| "learning_rate": 0.0001505778541741166, | |
| "loss": 0.3986, | |
| "step": 415 | |
| }, | |
| { | |
| "epoch": 1.883408071748879, | |
| "grad_norm": 0.341796875, | |
| "learning_rate": 0.00014794991516254793, | |
| "loss": 0.399, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 1.905829596412556, | |
| "grad_norm": 0.345703125, | |
| "learning_rate": 0.0001453426690025785, | |
| "loss": 0.402, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 1.9282511210762332, | |
| "grad_norm": 0.34765625, | |
| "learning_rate": 0.00014275779456466455, | |
| "loss": 0.4102, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 1.9506726457399104, | |
| "grad_norm": 0.32421875, | |
| "learning_rate": 0.00014019695631355567, | |
| "loss": 0.3942, | |
| "step": 435 | |
| }, | |
| { | |
| "epoch": 1.9730941704035874, | |
| "grad_norm": 0.333984375, | |
| "learning_rate": 0.0001376618032365048, | |
| "loss": 0.4017, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 1.9910313901345291, | |
| "eval_loss": 0.5547934770584106, | |
| "eval_runtime": 2.0196, | |
| "eval_samples_per_second": 20.796, | |
| "eval_steps_per_second": 20.796, | |
| "step": 444 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 669, | |
| "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": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 7.145669325918044e+17, | |
| "train_batch_size": 100, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |