| { | |
| "best_global_step": null, | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 10.0, | |
| "eval_steps": 500, | |
| "global_step": 640, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.015625, | |
| "grad_norm": 0.5563601851463318, | |
| "learning_rate": 3.125e-06, | |
| "loss": 1.837, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.078125, | |
| "grad_norm": 0.5579909682273865, | |
| "learning_rate": 1.5625e-05, | |
| "loss": 1.832, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.15625, | |
| "grad_norm": 0.3869417607784271, | |
| "learning_rate": 3.125e-05, | |
| "loss": 1.8211, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.234375, | |
| "grad_norm": 0.2597704827785492, | |
| "learning_rate": 4.6875e-05, | |
| "loss": 1.791, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.3125, | |
| "grad_norm": 0.2681010663509369, | |
| "learning_rate": 6.25e-05, | |
| "loss": 1.7632, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.390625, | |
| "grad_norm": 0.2458464354276657, | |
| "learning_rate": 7.8125e-05, | |
| "loss": 1.7142, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.46875, | |
| "grad_norm": 0.2320868819952011, | |
| "learning_rate": 9.375e-05, | |
| "loss": 1.662, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.546875, | |
| "grad_norm": 0.2217356562614441, | |
| "learning_rate": 0.000109375, | |
| "loss": 1.606, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.625, | |
| "grad_norm": 0.17081762850284576, | |
| "learning_rate": 0.000125, | |
| "loss": 1.5636, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.703125, | |
| "grad_norm": 0.13257570564746857, | |
| "learning_rate": 0.00014062500000000002, | |
| "loss": 1.5291, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.78125, | |
| "grad_norm": 0.12237544357776642, | |
| "learning_rate": 0.00015625, | |
| "loss": 1.5163, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.859375, | |
| "grad_norm": 0.10626661777496338, | |
| "learning_rate": 0.00017187500000000002, | |
| "loss": 1.4857, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.9375, | |
| "grad_norm": 0.1030210331082344, | |
| "learning_rate": 0.0001875, | |
| "loss": 1.471, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 2.3853657245635986, | |
| "eval_runtime": 0.8929, | |
| "eval_samples_per_second": 6.72, | |
| "eval_steps_per_second": 1.12, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 1.015625, | |
| "grad_norm": 0.1105547621846199, | |
| "learning_rate": 0.00019999851261394218, | |
| "loss": 1.4547, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 1.09375, | |
| "grad_norm": 0.1074226126074791, | |
| "learning_rate": 0.00019994645874763658, | |
| "loss": 1.4433, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.171875, | |
| "grad_norm": 0.1031486839056015, | |
| "learning_rate": 0.00019982007981886847, | |
| "loss": 1.4319, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "grad_norm": 0.09431572258472443, | |
| "learning_rate": 0.00019961946980917456, | |
| "loss": 1.4231, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.328125, | |
| "grad_norm": 0.10248947143554688, | |
| "learning_rate": 0.00019934477790194445, | |
| "loss": 1.414, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 1.40625, | |
| "grad_norm": 0.10952038317918777, | |
| "learning_rate": 0.00019899620837148077, | |
| "loss": 1.3994, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.484375, | |
| "grad_norm": 0.11935596913099289, | |
| "learning_rate": 0.0001985740204310909, | |
| "loss": 1.3933, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 1.5625, | |
| "grad_norm": 0.13777290284633636, | |
| "learning_rate": 0.00019807852804032305, | |
| "loss": 1.3891, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.640625, | |
| "grad_norm": 0.10964534431695938, | |
| "learning_rate": 0.00019751009967149087, | |
| "loss": 1.3755, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 1.71875, | |
| "grad_norm": 0.10717500746250153, | |
| "learning_rate": 0.00019686915803565934, | |
| "loss": 1.3811, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 1.796875, | |
| "grad_norm": 0.11473763734102249, | |
| "learning_rate": 0.0001961561797682962, | |
| "loss": 1.3661, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 1.875, | |
| "grad_norm": 0.11382485926151276, | |
| "learning_rate": 0.0001953716950748227, | |
| "loss": 1.3596, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.953125, | |
| "grad_norm": 0.12709280848503113, | |
| "learning_rate": 0.0001945162873363268, | |
| "loss": 1.362, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 2.375438690185547, | |
| "eval_runtime": 0.8905, | |
| "eval_samples_per_second": 6.738, | |
| "eval_steps_per_second": 1.123, | |
| "step": 128 | |
| }, | |
| { | |
| "epoch": 2.03125, | |
| "grad_norm": 0.1248418539762497, | |
| "learning_rate": 0.0001935905926757326, | |
| "loss": 1.3543, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 2.109375, | |
| "grad_norm": 0.13678818941116333, | |
| "learning_rate": 0.00019259529948474833, | |
| "loss": 1.3531, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 2.1875, | |
| "grad_norm": 0.12561410665512085, | |
| "learning_rate": 0.00019153114791194473, | |
| "loss": 1.3382, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 2.265625, | |
| "grad_norm": 0.14160117506980896, | |
| "learning_rate": 0.00019039892931234435, | |
| "loss": 1.3388, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 2.34375, | |
| "grad_norm": 0.1499018520116806, | |
| "learning_rate": 0.00018919948565893142, | |
| "loss": 1.3392, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 2.421875, | |
| "grad_norm": 0.14188511669635773, | |
| "learning_rate": 0.00018793370891651972, | |
| "loss": 1.3406, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "grad_norm": 0.12685342133045197, | |
| "learning_rate": 0.00018660254037844388, | |
| "loss": 1.3312, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 2.578125, | |
| "grad_norm": 0.1338503062725067, | |
| "learning_rate": 0.00018520696996656788, | |
| "loss": 1.3386, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 2.65625, | |
| "grad_norm": 0.1542833298444748, | |
| "learning_rate": 0.0001837480354951308, | |
| "loss": 1.3339, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 2.734375, | |
| "grad_norm": 0.1419745534658432, | |
| "learning_rate": 0.00018222682189897752, | |
| "loss": 1.3256, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 2.8125, | |
| "grad_norm": 0.12559981644153595, | |
| "learning_rate": 0.00018064446042674828, | |
| "loss": 1.3187, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 2.890625, | |
| "grad_norm": 0.1430642306804657, | |
| "learning_rate": 0.0001790021277996269, | |
| "loss": 1.3193, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 2.96875, | |
| "grad_norm": 0.12510241568088531, | |
| "learning_rate": 0.0001773010453362737, | |
| "loss": 1.3229, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 2.3739588260650635, | |
| "eval_runtime": 0.8887, | |
| "eval_samples_per_second": 6.751, | |
| "eval_steps_per_second": 1.125, | |
| "step": 192 | |
| }, | |
| { | |
| "epoch": 3.046875, | |
| "grad_norm": 0.13525685667991638, | |
| "learning_rate": 0.00017554247804459316, | |
| "loss": 1.3138, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 3.125, | |
| "grad_norm": 0.1221759021282196, | |
| "learning_rate": 0.0001737277336810124, | |
| "loss": 1.3114, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 3.203125, | |
| "grad_norm": 0.1508011817932129, | |
| "learning_rate": 0.0001718581617779698, | |
| "loss": 1.3121, | |
| "step": 205 | |
| }, | |
| { | |
| "epoch": 3.28125, | |
| "grad_norm": 0.15374642610549927, | |
| "learning_rate": 0.00016993515264033672, | |
| "loss": 1.3061, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 3.359375, | |
| "grad_norm": 0.13054175674915314, | |
| "learning_rate": 0.00016796013631151897, | |
| "loss": 1.3131, | |
| "step": 215 | |
| }, | |
| { | |
| "epoch": 3.4375, | |
| "grad_norm": 0.13617338240146637, | |
| "learning_rate": 0.00016593458151000688, | |
| "loss": 1.3055, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 3.515625, | |
| "grad_norm": 0.1574951857328415, | |
| "learning_rate": 0.00016385999453716454, | |
| "loss": 1.3081, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 3.59375, | |
| "grad_norm": 0.13046763837337494, | |
| "learning_rate": 0.00016173791815707051, | |
| "loss": 1.2971, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 3.671875, | |
| "grad_norm": 0.12328305840492249, | |
| "learning_rate": 0.00015956993044924334, | |
| "loss": 1.3004, | |
| "step": 235 | |
| }, | |
| { | |
| "epoch": 3.75, | |
| "grad_norm": 0.1457005739212036, | |
| "learning_rate": 0.0001573576436351046, | |
| "loss": 1.301, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 3.828125, | |
| "grad_norm": 0.14775606989860535, | |
| "learning_rate": 0.0001551027028790524, | |
| "loss": 1.2981, | |
| "step": 245 | |
| }, | |
| { | |
| "epoch": 3.90625, | |
| "grad_norm": 0.13573531806468964, | |
| "learning_rate": 0.0001528067850650368, | |
| "loss": 1.2964, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 3.984375, | |
| "grad_norm": 0.14599502086639404, | |
| "learning_rate": 0.0001504715975495472, | |
| "loss": 1.2996, | |
| "step": 255 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 2.3706400394439697, | |
| "eval_runtime": 0.8905, | |
| "eval_samples_per_second": 6.738, | |
| "eval_steps_per_second": 1.123, | |
| "step": 256 | |
| }, | |
| { | |
| "epoch": 4.0625, | |
| "grad_norm": 0.18352623283863068, | |
| "learning_rate": 0.00014809887689193877, | |
| "loss": 1.2791, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 4.140625, | |
| "grad_norm": 0.12809208035469055, | |
| "learning_rate": 0.00014569038756304207, | |
| "loss": 1.2838, | |
| "step": 265 | |
| }, | |
| { | |
| "epoch": 4.21875, | |
| "grad_norm": 0.14866453409194946, | |
| "learning_rate": 0.00014324792063301662, | |
| "loss": 1.2855, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 4.296875, | |
| "grad_norm": 0.1483401656150818, | |
| "learning_rate": 0.00014077329243942369, | |
| "loss": 1.2906, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 4.375, | |
| "grad_norm": 0.1301499307155609, | |
| "learning_rate": 0.000138268343236509, | |
| "loss": 1.2832, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 4.453125, | |
| "grad_norm": 0.12229609489440918, | |
| "learning_rate": 0.00013573493582670003, | |
| "loss": 1.2855, | |
| "step": 285 | |
| }, | |
| { | |
| "epoch": 4.53125, | |
| "grad_norm": 0.1406661719083786, | |
| "learning_rate": 0.00013317495417533524, | |
| "loss": 1.2885, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 4.609375, | |
| "grad_norm": 0.13113024830818176, | |
| "learning_rate": 0.00013059030200965536, | |
| "loss": 1.2899, | |
| "step": 295 | |
| }, | |
| { | |
| "epoch": 4.6875, | |
| "grad_norm": 0.13431058824062347, | |
| "learning_rate": 0.00012798290140309923, | |
| "loss": 1.2819, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 4.765625, | |
| "grad_norm": 0.14978773891925812, | |
| "learning_rate": 0.00012535469134595595, | |
| "loss": 1.2852, | |
| "step": 305 | |
| }, | |
| { | |
| "epoch": 4.84375, | |
| "grad_norm": 0.12423586845397949, | |
| "learning_rate": 0.00012270762630343734, | |
| "loss": 1.2802, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 4.921875, | |
| "grad_norm": 0.14971400797367096, | |
| "learning_rate": 0.00012004367476224206, | |
| "loss": 1.2829, | |
| "step": 315 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 0.14705318212509155, | |
| "learning_rate": 0.00011736481776669306, | |
| "loss": 1.2878, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 2.3755218982696533, | |
| "eval_runtime": 0.8842, | |
| "eval_samples_per_second": 6.786, | |
| "eval_steps_per_second": 1.131, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 5.078125, | |
| "grad_norm": 0.17323575913906097, | |
| "learning_rate": 0.00011467304744553618, | |
| "loss": 1.2791, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 5.15625, | |
| "grad_norm": 0.15446850657463074, | |
| "learning_rate": 0.00011197036553049625, | |
| "loss": 1.2663, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 5.234375, | |
| "grad_norm": 0.1357167363166809, | |
| "learning_rate": 0.00010925878186769158, | |
| "loss": 1.2707, | |
| "step": 335 | |
| }, | |
| { | |
| "epoch": 5.3125, | |
| "grad_norm": 0.13887612521648407, | |
| "learning_rate": 0.00010654031292301432, | |
| "loss": 1.2713, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 5.390625, | |
| "grad_norm": 0.18401511013507843, | |
| "learning_rate": 0.00010381698028258817, | |
| "loss": 1.2708, | |
| "step": 345 | |
| }, | |
| { | |
| "epoch": 5.46875, | |
| "grad_norm": 0.14264121651649475, | |
| "learning_rate": 0.00010109080914941824, | |
| "loss": 1.2716, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 5.546875, | |
| "grad_norm": 0.1242898479104042, | |
| "learning_rate": 9.836382683735132e-05, | |
| "loss": 1.2778, | |
| "step": 355 | |
| }, | |
| { | |
| "epoch": 5.625, | |
| "grad_norm": 0.11967791616916656, | |
| "learning_rate": 9.563806126346642e-05, | |
| "loss": 1.2761, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 5.703125, | |
| "grad_norm": 0.1407729834318161, | |
| "learning_rate": 9.29155394400166e-05, | |
| "loss": 1.271, | |
| "step": 365 | |
| }, | |
| { | |
| "epoch": 5.78125, | |
| "grad_norm": 0.127610981464386, | |
| "learning_rate": 9.019828596704394e-05, | |
| "loss": 1.2707, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 5.859375, | |
| "grad_norm": 0.131484255194664, | |
| "learning_rate": 8.74883215267881e-05, | |
| "loss": 1.2703, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 5.9375, | |
| "grad_norm": 0.14869672060012817, | |
| "learning_rate": 8.478766138100834e-05, | |
| "loss": 1.2746, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 2.3722288608551025, | |
| "eval_runtime": 0.8691, | |
| "eval_samples_per_second": 6.904, | |
| "eval_steps_per_second": 1.151, | |
| "step": 384 | |
| }, | |
| { | |
| "epoch": 6.015625, | |
| "grad_norm": 0.12939053773880005, | |
| "learning_rate": 8.209831387233676e-05, | |
| "loss": 1.267, | |
| "step": 385 | |
| }, | |
| { | |
| "epoch": 6.09375, | |
| "grad_norm": 0.1274917721748352, | |
| "learning_rate": 7.942227893077652e-05, | |
| "loss": 1.2609, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 6.171875, | |
| "grad_norm": 0.128373920917511, | |
| "learning_rate": 7.676154658645656e-05, | |
| "loss": 1.2613, | |
| "step": 395 | |
| }, | |
| { | |
| "epoch": 6.25, | |
| "grad_norm": 0.12212257087230682, | |
| "learning_rate": 7.411809548974792e-05, | |
| "loss": 1.2608, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 6.328125, | |
| "grad_norm": 0.12020603567361832, | |
| "learning_rate": 7.149389143984295e-05, | |
| "loss": 1.2648, | |
| "step": 405 | |
| }, | |
| { | |
| "epoch": 6.40625, | |
| "grad_norm": 0.13269896805286407, | |
| "learning_rate": 6.889088592289093e-05, | |
| "loss": 1.263, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 6.484375, | |
| "grad_norm": 0.12962587177753448, | |
| "learning_rate": 6.6311014660778e-05, | |
| "loss": 1.2688, | |
| "step": 415 | |
| }, | |
| { | |
| "epoch": 6.5625, | |
| "grad_norm": 0.13470596075057983, | |
| "learning_rate": 6.375619617162985e-05, | |
| "loss": 1.2606, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 6.640625, | |
| "grad_norm": 0.1306695193052292, | |
| "learning_rate": 6.122833034310793e-05, | |
| "loss": 1.264, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 6.71875, | |
| "grad_norm": 0.12205986678600311, | |
| "learning_rate": 5.872929701956054e-05, | |
| "loss": 1.2624, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 6.796875, | |
| "grad_norm": 0.12223726511001587, | |
| "learning_rate": 5.6260954604078585e-05, | |
| "loss": 1.2624, | |
| "step": 435 | |
| }, | |
| { | |
| "epoch": 6.875, | |
| "grad_norm": 0.11978649348020554, | |
| "learning_rate": 5.382513867649663e-05, | |
| "loss": 1.2684, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 6.953125, | |
| "grad_norm": 0.13118350505828857, | |
| "learning_rate": 5.142366062836599e-05, | |
| "loss": 1.2617, | |
| "step": 445 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_loss": 2.3756062984466553, | |
| "eval_runtime": 0.8698, | |
| "eval_samples_per_second": 6.898, | |
| "eval_steps_per_second": 1.15, | |
| "step": 448 | |
| }, | |
| { | |
| "epoch": 7.03125, | |
| "grad_norm": 0.12501190602779388, | |
| "learning_rate": 4.9058306315915826e-05, | |
| "loss": 1.2686, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 7.109375, | |
| "grad_norm": 0.11916686594486237, | |
| "learning_rate": 4.6730834732003104e-05, | |
| "loss": 1.2589, | |
| "step": 455 | |
| }, | |
| { | |
| "epoch": 7.1875, | |
| "grad_norm": 0.1307828575372696, | |
| "learning_rate": 4.444297669803981e-05, | |
| "loss": 1.2641, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 7.265625, | |
| "grad_norm": 0.1322871744632721, | |
| "learning_rate": 4.219643357686967e-05, | |
| "loss": 1.2637, | |
| "step": 465 | |
| }, | |
| { | |
| "epoch": 7.34375, | |
| "grad_norm": 0.12456091493368149, | |
| "learning_rate": 3.999287600755192e-05, | |
| "loss": 1.2596, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 7.421875, | |
| "grad_norm": 0.12441947311162949, | |
| "learning_rate": 3.783394266299228e-05, | |
| "loss": 1.2609, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 7.5, | |
| "grad_norm": 0.1233881339430809, | |
| "learning_rate": 3.5721239031346066e-05, | |
| "loss": 1.258, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 7.578125, | |
| "grad_norm": 0.11959797888994217, | |
| "learning_rate": 3.365633622209891e-05, | |
| "loss": 1.2535, | |
| "step": 485 | |
| }, | |
| { | |
| "epoch": 7.65625, | |
| "grad_norm": 0.13552451133728027, | |
| "learning_rate": 3.164076979771287e-05, | |
| "loss": 1.2567, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 7.734375, | |
| "grad_norm": 0.1312321126461029, | |
| "learning_rate": 2.9676038631707593e-05, | |
| "loss": 1.2546, | |
| "step": 495 | |
| }, | |
| { | |
| "epoch": 7.8125, | |
| "grad_norm": 0.12341451644897461, | |
| "learning_rate": 2.776360379402445e-05, | |
| "loss": 1.2589, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 7.890625, | |
| "grad_norm": 0.11580604314804077, | |
| "learning_rate": 2.5904887464504114e-05, | |
| "loss": 1.2514, | |
| "step": 505 | |
| }, | |
| { | |
| "epoch": 7.96875, | |
| "grad_norm": 0.1164306178689003, | |
| "learning_rate": 2.4101271875283817e-05, | |
| "loss": 1.2497, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 2.3754327297210693, | |
| "eval_runtime": 0.8715, | |
| "eval_samples_per_second": 6.885, | |
| "eval_steps_per_second": 1.147, | |
| "step": 512 | |
| }, | |
| { | |
| "epoch": 8.046875, | |
| "grad_norm": 0.11353053152561188, | |
| "learning_rate": 2.2354098282902446e-05, | |
| "loss": 1.2515, | |
| "step": 515 | |
| }, | |
| { | |
| "epoch": 8.125, | |
| "grad_norm": 0.11808107793331146, | |
| "learning_rate": 2.0664665970876496e-05, | |
| "loss": 1.2568, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 8.203125, | |
| "grad_norm": 0.12227106839418411, | |
| "learning_rate": 1.903423128348959e-05, | |
| "loss": 1.2526, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 8.28125, | |
| "grad_norm": 0.11858697980642319, | |
| "learning_rate": 1.7464006691513623e-05, | |
| "loss": 1.2546, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 8.359375, | |
| "grad_norm": 0.11320216953754425, | |
| "learning_rate": 1.595515989055618e-05, | |
| "loss": 1.2513, | |
| "step": 535 | |
| }, | |
| { | |
| "epoch": 8.4375, | |
| "grad_norm": 0.11540042608976364, | |
| "learning_rate": 1.4508812932705363e-05, | |
| "loss": 1.2539, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 8.515625, | |
| "grad_norm": 0.11582642793655396, | |
| "learning_rate": 1.3126041392116772e-05, | |
| "loss": 1.2552, | |
| "step": 545 | |
| }, | |
| { | |
| "epoch": 8.59375, | |
| "grad_norm": 0.11406353861093521, | |
| "learning_rate": 1.1807873565164506e-05, | |
| "loss": 1.2451, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 8.671875, | |
| "grad_norm": 0.11722878366708755, | |
| "learning_rate": 1.0555289705749483e-05, | |
| "loss": 1.2555, | |
| "step": 555 | |
| }, | |
| { | |
| "epoch": 8.75, | |
| "grad_norm": 0.11788377165794373, | |
| "learning_rate": 9.369221296335006e-06, | |
| "loss": 1.2644, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 8.828125, | |
| "grad_norm": 0.12228409945964813, | |
| "learning_rate": 8.250550355250875e-06, | |
| "loss": 1.2554, | |
| "step": 565 | |
| }, | |
| { | |
| "epoch": 8.90625, | |
| "grad_norm": 0.11590871214866638, | |
| "learning_rate": 7.200108780781556e-06, | |
| "loss": 1.2581, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 8.984375, | |
| "grad_norm": 0.11986227333545685, | |
| "learning_rate": 6.218677732526035e-06, | |
| "loss": 1.2549, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_loss": 2.3762052059173584, | |
| "eval_runtime": 0.8701, | |
| "eval_samples_per_second": 6.896, | |
| "eval_steps_per_second": 1.149, | |
| "step": 576 | |
| }, | |
| { | |
| "epoch": 9.0625, | |
| "grad_norm": 0.11538632959127426, | |
| "learning_rate": 5.306987050489442e-06, | |
| "loss": 1.2553, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 9.140625, | |
| "grad_norm": 0.11359097808599472, | |
| "learning_rate": 4.465714712338398e-06, | |
| "loss": 1.2438, | |
| "step": 585 | |
| }, | |
| { | |
| "epoch": 9.21875, | |
| "grad_norm": 0.1167021319270134, | |
| "learning_rate": 3.6954863292237297e-06, | |
| "loss": 1.2508, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 9.296875, | |
| "grad_norm": 0.11401596665382385, | |
| "learning_rate": 2.996874680545603e-06, | |
| "loss": 1.2555, | |
| "step": 595 | |
| }, | |
| { | |
| "epoch": 9.375, | |
| "grad_norm": 0.11593978106975555, | |
| "learning_rate": 2.3703992880066638e-06, | |
| "loss": 1.2491, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 9.453125, | |
| "grad_norm": 0.11369957774877548, | |
| "learning_rate": 1.8165260292704711e-06, | |
| "loss": 1.2556, | |
| "step": 605 | |
| }, | |
| { | |
| "epoch": 9.53125, | |
| "grad_norm": 0.11447325348854065, | |
| "learning_rate": 1.3356667915121025e-06, | |
| "loss": 1.2559, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 9.609375, | |
| "grad_norm": 0.11217474192380905, | |
| "learning_rate": 9.281791651187366e-07, | |
| "loss": 1.2635, | |
| "step": 615 | |
| }, | |
| { | |
| "epoch": 9.6875, | |
| "grad_norm": 0.11821179836988449, | |
| "learning_rate": 5.943661777680354e-07, | |
| "loss": 1.2521, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 9.765625, | |
| "grad_norm": 0.113133005797863, | |
| "learning_rate": 3.3447606908196817e-07, | |
| "loss": 1.2539, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 9.84375, | |
| "grad_norm": 0.11367449164390564, | |
| "learning_rate": 1.487021060236904e-07, | |
| "loss": 1.2528, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 9.921875, | |
| "grad_norm": 0.11390256881713867, | |
| "learning_rate": 3.7182439174832106e-08, | |
| "loss": 1.2581, | |
| "step": 635 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 0.11366365104913712, | |
| "learning_rate": 0.0, | |
| "loss": 1.2494, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_loss": 2.376882791519165, | |
| "eval_runtime": 0.8974, | |
| "eval_samples_per_second": 6.686, | |
| "eval_steps_per_second": 1.114, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "step": 640, | |
| "total_flos": 2.846679860191953e+18, | |
| "train_loss": 1.3252380434423685, | |
| "train_runtime": 3360.483, | |
| "train_samples_per_second": 48.669, | |
| "train_steps_per_second": 0.19 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 640, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 100, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2.846679860191953e+18, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |