| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.9994358251057829, | |
| "eval_steps": 500, | |
| "global_step": 886, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.022566995768688293, | |
| "grad_norm": 1.1076090467924995, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7438, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.045133991537376586, | |
| "grad_norm": 0.991735987363898, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6802, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.06770098730606489, | |
| "grad_norm": 0.8721792478756093, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6629, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.09026798307475317, | |
| "grad_norm": 0.5951545166641695, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6406, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.11283497884344147, | |
| "grad_norm": 0.7007845996565958, | |
| "learning_rate": 5e-06, | |
| "loss": 0.634, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.13540197461212977, | |
| "grad_norm": 0.5943811510802003, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6332, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.15796897038081806, | |
| "grad_norm": 0.6353234676535418, | |
| "learning_rate": 5e-06, | |
| "loss": 0.629, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.18053596614950634, | |
| "grad_norm": 0.6006567013207602, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6306, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.20310296191819463, | |
| "grad_norm": 0.7046632924461987, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6282, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.22566995768688294, | |
| "grad_norm": 0.7021240097891023, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6272, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.24823695345557123, | |
| "grad_norm": 0.5910229174924321, | |
| "learning_rate": 5e-06, | |
| "loss": 0.618, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.27080394922425954, | |
| "grad_norm": 0.596959901149353, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6173, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.2933709449929478, | |
| "grad_norm": 0.8993218759485339, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6185, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.3159379407616361, | |
| "grad_norm": 0.6147671660802024, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6135, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.3385049365303244, | |
| "grad_norm": 0.6724362951434409, | |
| "learning_rate": 5e-06, | |
| "loss": 0.614, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.3610719322990127, | |
| "grad_norm": 0.6020059058264652, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6157, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.383638928067701, | |
| "grad_norm": 0.6467062739486626, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6118, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.40620592383638926, | |
| "grad_norm": 0.6337134461733973, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6087, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.4287729196050776, | |
| "grad_norm": 1.187290343191463, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6032, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.4513399153737659, | |
| "grad_norm": 0.6573468292913784, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6099, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.47390691114245415, | |
| "grad_norm": 0.5436088131885063, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6063, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.49647390691114246, | |
| "grad_norm": 0.5714729478422241, | |
| "learning_rate": 5e-06, | |
| "loss": 0.598, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.5190409026798307, | |
| "grad_norm": 0.6369923617172581, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6068, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.5416078984485191, | |
| "grad_norm": 0.6174803589160689, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6057, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.5641748942172073, | |
| "grad_norm": 0.5746679987420106, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5998, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.5867418899858956, | |
| "grad_norm": 0.7158022714822785, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6025, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.609308885754584, | |
| "grad_norm": 0.6550177518697455, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5979, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.6318758815232722, | |
| "grad_norm": 0.5848914985146456, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6045, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.6544428772919605, | |
| "grad_norm": 0.6013598059389905, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5979, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.6770098730606487, | |
| "grad_norm": 0.6694540638792662, | |
| "learning_rate": 5e-06, | |
| "loss": 0.604, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.6995768688293371, | |
| "grad_norm": 0.5419093913651906, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5972, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.7221438645980254, | |
| "grad_norm": 0.5432682430557793, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6004, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.7447108603667136, | |
| "grad_norm": 0.5609209464002083, | |
| "learning_rate": 5e-06, | |
| "loss": 0.596, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.767277856135402, | |
| "grad_norm": 0.5771320670008115, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5955, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.7898448519040903, | |
| "grad_norm": 0.6553502516133991, | |
| "learning_rate": 5e-06, | |
| "loss": 0.597, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.8124118476727785, | |
| "grad_norm": 0.5735610527819333, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5959, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.8349788434414669, | |
| "grad_norm": 0.6196981942326923, | |
| "learning_rate": 5e-06, | |
| "loss": 0.603, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.8575458392101551, | |
| "grad_norm": 0.5892757008227134, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5997, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.8801128349788434, | |
| "grad_norm": 0.6377051547834243, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5937, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.9026798307475318, | |
| "grad_norm": 0.5793619434689786, | |
| "learning_rate": 5e-06, | |
| "loss": 0.596, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.92524682651622, | |
| "grad_norm": 0.5940412617035421, | |
| "learning_rate": 5e-06, | |
| "loss": 0.594, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 0.9478138222849083, | |
| "grad_norm": 0.5720643197976489, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5946, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.9703808180535967, | |
| "grad_norm": 0.5934993591499655, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5886, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 0.9929478138222849, | |
| "grad_norm": 0.7147785717960933, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5934, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 0.9997179125528914, | |
| "eval_loss": 0.5912777781486511, | |
| "eval_runtime": 686.2871, | |
| "eval_samples_per_second": 17.399, | |
| "eval_steps_per_second": 0.545, | |
| "step": 443 | |
| }, | |
| { | |
| "epoch": 1.0155148095909732, | |
| "grad_norm": 0.6953054875114835, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6008, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 1.0380818053596614, | |
| "grad_norm": 0.5903993048531896, | |
| "learning_rate": 5e-06, | |
| "loss": 0.528, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 1.0606488011283497, | |
| "grad_norm": 0.6324628267144559, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5429, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 1.0832157968970382, | |
| "grad_norm": 0.5764061185246054, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5308, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 1.1057827926657264, | |
| "grad_norm": 0.6012235908025508, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5408, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 1.1283497884344147, | |
| "grad_norm": 0.5789335395759082, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5396, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.150916784203103, | |
| "grad_norm": 0.6109006953078554, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5407, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 1.1734837799717912, | |
| "grad_norm": 0.5684452419181554, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5333, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 1.1960507757404795, | |
| "grad_norm": 0.6963518157721161, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5395, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 1.2186177715091677, | |
| "grad_norm": 0.6141353364563702, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5371, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 1.2411847672778562, | |
| "grad_norm": 0.6539340543845323, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5419, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 1.2637517630465445, | |
| "grad_norm": 0.6168349027651291, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5438, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 1.2863187588152327, | |
| "grad_norm": 0.581736789269481, | |
| "learning_rate": 5e-06, | |
| "loss": 0.541, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 1.308885754583921, | |
| "grad_norm": 0.5594406044592434, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5409, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 1.3314527503526092, | |
| "grad_norm": 0.5620275131748222, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5411, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 1.3540197461212977, | |
| "grad_norm": 0.6238286913601226, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5463, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.376586741889986, | |
| "grad_norm": 0.5415008064753611, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5391, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 1.3991537376586742, | |
| "grad_norm": 0.5610560374369729, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5433, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 1.4217207334273625, | |
| "grad_norm": 0.550463676620283, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5413, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 1.4442877291960508, | |
| "grad_norm": 0.639666171350872, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5409, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 1.466854724964739, | |
| "grad_norm": 0.5585935410492179, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5418, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 1.4894217207334273, | |
| "grad_norm": 0.6134871404266905, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5456, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 1.5119887165021155, | |
| "grad_norm": 0.5682459714599422, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5403, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 1.5345557122708038, | |
| "grad_norm": 0.5522175532246568, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5344, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 1.5571227080394923, | |
| "grad_norm": 0.5925815010766872, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5366, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 1.5796897038081805, | |
| "grad_norm": 0.6578651642561647, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5408, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.6022566995768688, | |
| "grad_norm": 0.5408548916824767, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5435, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 1.6248236953455573, | |
| "grad_norm": 0.5988342872882225, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5518, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 1.6473906911142455, | |
| "grad_norm": 0.5384009297069935, | |
| "learning_rate": 5e-06, | |
| "loss": 0.535, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 1.6699576868829338, | |
| "grad_norm": 0.7024106203910366, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5508, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 1.692524682651622, | |
| "grad_norm": 0.6637725967890132, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5422, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 1.7150916784203103, | |
| "grad_norm": 0.6071900872105588, | |
| "learning_rate": 5e-06, | |
| "loss": 0.541, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 1.7376586741889986, | |
| "grad_norm": 0.6565272249072508, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5417, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 1.7602256699576868, | |
| "grad_norm": 0.5523582207595168, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5375, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 1.782792665726375, | |
| "grad_norm": 0.5699161494100724, | |
| "learning_rate": 5e-06, | |
| "loss": 0.541, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 1.8053596614950633, | |
| "grad_norm": 0.5993412920248554, | |
| "learning_rate": 5e-06, | |
| "loss": 0.536, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.8279266572637518, | |
| "grad_norm": 0.7661220656377966, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5468, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 1.85049365303244, | |
| "grad_norm": 0.6803479582333226, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5341, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 1.8730606488011283, | |
| "grad_norm": 0.5645527605654231, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5462, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 1.8956276445698168, | |
| "grad_norm": 0.6297088212997327, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5461, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 1.918194640338505, | |
| "grad_norm": 0.555460659706579, | |
| "learning_rate": 5e-06, | |
| "loss": 0.534, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 1.9407616361071933, | |
| "grad_norm": 0.6152170831362264, | |
| "learning_rate": 5e-06, | |
| "loss": 0.539, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 1.9633286318758816, | |
| "grad_norm": 0.8147767450293179, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5456, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 1.9858956276445698, | |
| "grad_norm": 0.5537465355114557, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5439, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 1.9994358251057829, | |
| "eval_loss": 0.5872690677642822, | |
| "eval_runtime": 682.0272, | |
| "eval_samples_per_second": 17.508, | |
| "eval_steps_per_second": 0.548, | |
| "step": 886 | |
| }, | |
| { | |
| "epoch": 1.9994358251057829, | |
| "step": 886, | |
| "total_flos": 3376050654412800.0, | |
| "train_loss": 0.5777317097170896, | |
| "train_runtime": 80301.89, | |
| "train_samples_per_second": 5.65, | |
| "train_steps_per_second": 0.011 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 886, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 2, | |
| "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": 3376050654412800.0, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |