| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.999438727782975, | |
| "eval_steps": 500, | |
| "global_step": 1002, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.029934518241347054, | |
| "grad_norm": 1.3711546008530633, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7728, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.05986903648269411, | |
| "grad_norm": 0.8532931970235523, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6958, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.08980355472404115, | |
| "grad_norm": 0.9311714188883703, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6702, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.11973807296538821, | |
| "grad_norm": 0.7891322057145671, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6661, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.14967259120673526, | |
| "grad_norm": 0.6667175607439161, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6572, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.1796071094480823, | |
| "grad_norm": 0.6870807802747252, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6491, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.20954162768942938, | |
| "grad_norm": 0.7923438547434251, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6458, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.23947614593077643, | |
| "grad_norm": 0.7026808650849072, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6473, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.2694106641721235, | |
| "grad_norm": 0.9113469922865229, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6385, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.2993451824134705, | |
| "grad_norm": 0.6673532156562788, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6397, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.3292797006548176, | |
| "grad_norm": 0.7024293184544007, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6355, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.3592142188961646, | |
| "grad_norm": 0.7252389528859061, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6343, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.3891487371375117, | |
| "grad_norm": 0.7166274742588952, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6345, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.41908325537885877, | |
| "grad_norm": 1.044790711824065, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6283, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.4490177736202058, | |
| "grad_norm": 0.6941559597539316, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6299, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.47895229186155286, | |
| "grad_norm": 0.6459395559042749, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6287, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.5088868101028999, | |
| "grad_norm": 0.8200206082293998, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6231, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.538821328344247, | |
| "grad_norm": 0.8147867268820529, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6247, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.568755846585594, | |
| "grad_norm": 0.7086057878705686, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6225, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.598690364826941, | |
| "grad_norm": 0.6936168023515021, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6302, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.6286248830682881, | |
| "grad_norm": 0.7668440736078843, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6242, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.6585594013096352, | |
| "grad_norm": 0.9231622196440751, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6223, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.6884939195509823, | |
| "grad_norm": 0.7268095301330812, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6248, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.7184284377923292, | |
| "grad_norm": 0.7253911527197313, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6237, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.7483629560336763, | |
| "grad_norm": 0.624367345766973, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6202, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.7782974742750234, | |
| "grad_norm": 0.7459567304697926, | |
| "learning_rate": 5e-06, | |
| "loss": 0.619, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.8082319925163705, | |
| "grad_norm": 0.7173470697470193, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6264, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.8381665107577175, | |
| "grad_norm": 0.6676024114471235, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6165, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.8681010289990645, | |
| "grad_norm": 0.7422170215057577, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6133, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.8980355472404116, | |
| "grad_norm": 0.770476706070457, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6147, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.9279700654817586, | |
| "grad_norm": 0.7280386561072427, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6135, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.9579045837231057, | |
| "grad_norm": 0.6377527403933052, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6155, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.9878391019644528, | |
| "grad_norm": 0.9160036426859673, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6084, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.9998129092609915, | |
| "eval_loss": 0.6195828914642334, | |
| "eval_runtime": 514.5322, | |
| "eval_samples_per_second": 17.494, | |
| "eval_steps_per_second": 0.548, | |
| "step": 334 | |
| }, | |
| { | |
| "epoch": 1.0177736202057999, | |
| "grad_norm": 0.8698832997202014, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6386, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 1.047708138447147, | |
| "grad_norm": 0.818554628982315, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5525, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.077642656688494, | |
| "grad_norm": 0.7038909119697851, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5481, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 1.1075771749298409, | |
| "grad_norm": 0.7417662574743294, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5481, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 1.137511693171188, | |
| "grad_norm": 0.6971090185954912, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5483, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 1.167446211412535, | |
| "grad_norm": 0.6877839912070208, | |
| "learning_rate": 5e-06, | |
| "loss": 0.55, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 1.197380729653882, | |
| "grad_norm": 0.7726546070457438, | |
| "learning_rate": 5e-06, | |
| "loss": 0.55, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.2273152478952292, | |
| "grad_norm": 0.627193492530289, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5524, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 1.2572497661365762, | |
| "grad_norm": 0.7305679948147004, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5567, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 1.2871842843779233, | |
| "grad_norm": 0.6901834732706111, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5513, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 1.3171188026192704, | |
| "grad_norm": 0.7078951488952088, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5533, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 1.3470533208606175, | |
| "grad_norm": 0.6469932896858731, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5474, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 1.3769878391019645, | |
| "grad_norm": 0.6984755612841984, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5579, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 1.4069223573433116, | |
| "grad_norm": 0.7417844126416642, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5571, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 1.4368568755846587, | |
| "grad_norm": 0.717182519772729, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5591, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 1.4667913938260055, | |
| "grad_norm": 0.6313303871675323, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5585, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 1.4967259120673526, | |
| "grad_norm": 0.7084037752703477, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5546, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.5266604303086997, | |
| "grad_norm": 0.6531206728146401, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5577, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 1.5565949485500468, | |
| "grad_norm": 0.654485495629501, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5555, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 1.5865294667913938, | |
| "grad_norm": 0.6725661891430049, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5561, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 1.616463985032741, | |
| "grad_norm": 0.698329402386774, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5599, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 1.646398503274088, | |
| "grad_norm": 0.6690114035686613, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5592, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 1.6763330215154348, | |
| "grad_norm": 0.66011999214286, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5498, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 1.706267539756782, | |
| "grad_norm": 0.8040306195233741, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5518, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 1.736202057998129, | |
| "grad_norm": 0.7060300162023749, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5529, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 1.766136576239476, | |
| "grad_norm": 0.6673011361622929, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5558, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 1.7960710944808231, | |
| "grad_norm": 0.7571655613304361, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5564, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.8260056127221702, | |
| "grad_norm": 0.642019011432012, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5543, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 1.8559401309635173, | |
| "grad_norm": 0.6324305634532394, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5524, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 1.8858746492048644, | |
| "grad_norm": 0.6741301602507832, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5543, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 1.9158091674462114, | |
| "grad_norm": 0.7541104518831577, | |
| "learning_rate": 5e-06, | |
| "loss": 0.556, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 1.9457436856875585, | |
| "grad_norm": 0.6493122341038763, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5541, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 1.9756782039289056, | |
| "grad_norm": 0.8173207800207763, | |
| "learning_rate": 5e-06, | |
| "loss": 0.559, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 1.999625818521983, | |
| "eval_loss": 0.6155872941017151, | |
| "eval_runtime": 513.9247, | |
| "eval_samples_per_second": 17.514, | |
| "eval_steps_per_second": 0.549, | |
| "step": 668 | |
| }, | |
| { | |
| "epoch": 2.0056127221702527, | |
| "grad_norm": 1.2127948243552868, | |
| "learning_rate": 5e-06, | |
| "loss": 0.6011, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 2.0355472404115997, | |
| "grad_norm": 0.8183124334466546, | |
| "learning_rate": 5e-06, | |
| "loss": 0.4861, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 2.065481758652947, | |
| "grad_norm": 0.7278809818448668, | |
| "learning_rate": 5e-06, | |
| "loss": 0.4822, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 2.095416276894294, | |
| "grad_norm": 0.6907315590571528, | |
| "learning_rate": 5e-06, | |
| "loss": 0.4866, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 2.125350795135641, | |
| "grad_norm": 0.7350401872049226, | |
| "learning_rate": 5e-06, | |
| "loss": 0.4864, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 2.155285313376988, | |
| "grad_norm": 0.8001247911358127, | |
| "learning_rate": 5e-06, | |
| "loss": 0.4898, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 2.185219831618335, | |
| "grad_norm": 0.7054933220563573, | |
| "learning_rate": 5e-06, | |
| "loss": 0.4897, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 2.2151543498596817, | |
| "grad_norm": 0.7802502700344892, | |
| "learning_rate": 5e-06, | |
| "loss": 0.4901, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 2.245088868101029, | |
| "grad_norm": 0.7215036811131982, | |
| "learning_rate": 5e-06, | |
| "loss": 0.4923, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 2.275023386342376, | |
| "grad_norm": 0.6977267806305402, | |
| "learning_rate": 5e-06, | |
| "loss": 0.4878, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 2.304957904583723, | |
| "grad_norm": 0.7014967299126638, | |
| "learning_rate": 5e-06, | |
| "loss": 0.4929, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 2.33489242282507, | |
| "grad_norm": 0.6887436747500622, | |
| "learning_rate": 5e-06, | |
| "loss": 0.4887, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 2.364826941066417, | |
| "grad_norm": 0.6792765123813863, | |
| "learning_rate": 5e-06, | |
| "loss": 0.4905, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 2.394761459307764, | |
| "grad_norm": 0.7699608340540777, | |
| "learning_rate": 5e-06, | |
| "loss": 0.4933, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.4246959775491113, | |
| "grad_norm": 0.7845823284512902, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5002, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 2.4546304957904583, | |
| "grad_norm": 0.7943786395213068, | |
| "learning_rate": 5e-06, | |
| "loss": 0.4973, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 2.4845650140318054, | |
| "grad_norm": 0.7944786954591401, | |
| "learning_rate": 5e-06, | |
| "loss": 0.497, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 2.5144995322731525, | |
| "grad_norm": 0.7339137535531504, | |
| "learning_rate": 5e-06, | |
| "loss": 0.4944, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 2.5444340505144996, | |
| "grad_norm": 0.6904479862339539, | |
| "learning_rate": 5e-06, | |
| "loss": 0.4978, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 2.5743685687558466, | |
| "grad_norm": 0.7239438693443128, | |
| "learning_rate": 5e-06, | |
| "loss": 0.496, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 2.6043030869971937, | |
| "grad_norm": 0.6527578782432896, | |
| "learning_rate": 5e-06, | |
| "loss": 0.4936, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 2.634237605238541, | |
| "grad_norm": 0.7144829092908652, | |
| "learning_rate": 5e-06, | |
| "loss": 0.492, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 2.664172123479888, | |
| "grad_norm": 0.7507359205267485, | |
| "learning_rate": 5e-06, | |
| "loss": 0.498, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 2.694106641721235, | |
| "grad_norm": 0.698227000026423, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5006, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 2.724041159962582, | |
| "grad_norm": 0.6978873719386385, | |
| "learning_rate": 5e-06, | |
| "loss": 0.4942, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 2.753975678203929, | |
| "grad_norm": 0.6619027440664284, | |
| "learning_rate": 5e-06, | |
| "loss": 0.4955, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 2.7839101964452757, | |
| "grad_norm": 0.6892171519418343, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5013, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 2.8138447146866232, | |
| "grad_norm": 0.7177100967350676, | |
| "learning_rate": 5e-06, | |
| "loss": 0.4994, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 2.84377923292797, | |
| "grad_norm": 0.7555576282537906, | |
| "learning_rate": 5e-06, | |
| "loss": 0.4988, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 2.8737137511693174, | |
| "grad_norm": 0.8751247226062729, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5009, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 2.903648269410664, | |
| "grad_norm": 0.730851224108063, | |
| "learning_rate": 5e-06, | |
| "loss": 0.4964, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 2.933582787652011, | |
| "grad_norm": 0.7044959813969391, | |
| "learning_rate": 5e-06, | |
| "loss": 0.502, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 2.963517305893358, | |
| "grad_norm": 0.7404990276858111, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5036, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 2.9934518241347052, | |
| "grad_norm": 0.7940138869090708, | |
| "learning_rate": 5e-06, | |
| "loss": 0.5047, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.999438727782975, | |
| "eval_loss": 0.6338862776756287, | |
| "eval_runtime": 513.4832, | |
| "eval_samples_per_second": 17.529, | |
| "eval_steps_per_second": 0.549, | |
| "step": 1002 | |
| }, | |
| { | |
| "epoch": 2.999438727782975, | |
| "step": 1002, | |
| "total_flos": 3818092983484416.0, | |
| "train_loss": 0.5629288574416719, | |
| "train_runtime": 90608.8212, | |
| "train_samples_per_second": 5.662, | |
| "train_steps_per_second": 0.011 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 1002, | |
| "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": 3818092983484416.0, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |