{ "best_metric": null, "best_model_checkpoint": null, "epoch": 2.9991537376586743, "eval_steps": 500, "global_step": 1329, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.022566995768688293, "grad_norm": 2.863170566955824, "learning_rate": 7.462686567164179e-07, "loss": 0.8503, "step": 10 }, { "epoch": 0.045133991537376586, "grad_norm": 1.0142792486813195, "learning_rate": 1.4925373134328358e-06, "loss": 0.7616, "step": 20 }, { "epoch": 0.06770098730606489, "grad_norm": 0.8684906705560901, "learning_rate": 2.238805970149254e-06, "loss": 0.7089, "step": 30 }, { "epoch": 0.09026798307475317, "grad_norm": 0.8738166161325526, "learning_rate": 2.9850746268656716e-06, "loss": 0.6699, "step": 40 }, { "epoch": 0.11283497884344147, "grad_norm": 0.68350941569959, "learning_rate": 3.73134328358209e-06, "loss": 0.6543, "step": 50 }, { "epoch": 0.13540197461212977, "grad_norm": 0.6782468417382832, "learning_rate": 4.477611940298508e-06, "loss": 0.6478, "step": 60 }, { "epoch": 0.15796897038081806, "grad_norm": 1.224732485553396, "learning_rate": 4.999937255733162e-06, "loss": 0.6409, "step": 70 }, { "epoch": 0.18053596614950634, "grad_norm": 0.6451209109161937, "learning_rate": 4.9988218994478215e-06, "loss": 0.6408, "step": 80 }, { "epoch": 0.20310296191819463, "grad_norm": 0.7237779324582321, "learning_rate": 4.9963130216752534e-06, "loss": 0.6369, "step": 90 }, { "epoch": 0.22566995768688294, "grad_norm": 0.6251777291331431, "learning_rate": 4.992412177085074e-06, "loss": 0.6342, "step": 100 }, { "epoch": 0.24823695345557123, "grad_norm": 0.677156166674141, "learning_rate": 4.98712178290328e-06, "loss": 0.6243, "step": 110 }, { "epoch": 0.27080394922425954, "grad_norm": 0.6822617228770884, "learning_rate": 4.980445117414368e-06, "loss": 0.623, "step": 120 }, { "epoch": 0.2933709449929478, "grad_norm": 0.8294207021685933, "learning_rate": 4.97238631792989e-06, "loss": 0.6238, "step": 130 }, { "epoch": 0.3159379407616361, "grad_norm": 0.7385138099481724, "learning_rate": 4.962950378224695e-06, "loss": 0.6184, "step": 140 }, { "epoch": 0.3385049365303244, "grad_norm": 0.657291551157545, "learning_rate": 4.9521431454424524e-06, "loss": 0.6185, "step": 150 }, { "epoch": 0.3610719322990127, "grad_norm": 0.6296103452981768, "learning_rate": 4.939971316472365e-06, "loss": 0.6203, "step": 160 }, { "epoch": 0.383638928067701, "grad_norm": 0.6567437618009186, "learning_rate": 4.926442433799328e-06, "loss": 0.6159, "step": 170 }, { "epoch": 0.40620592383638926, "grad_norm": 0.6118505861612987, "learning_rate": 4.911564880830097e-06, "loss": 0.6123, "step": 180 }, { "epoch": 0.4287729196050776, "grad_norm": 0.608225756194811, "learning_rate": 4.895347876698363e-06, "loss": 0.6067, "step": 190 }, { "epoch": 0.4513399153737659, "grad_norm": 0.6416769574175134, "learning_rate": 4.877801470551962e-06, "loss": 0.6129, "step": 200 }, { "epoch": 0.47390691114245415, "grad_norm": 0.5757042759804658, "learning_rate": 4.858936535325733e-06, "loss": 0.6092, "step": 210 }, { "epoch": 0.49647390691114246, "grad_norm": 0.6950103834112719, "learning_rate": 4.838764761003932e-06, "loss": 0.6008, "step": 220 }, { "epoch": 0.5190409026798307, "grad_norm": 0.6120991974648448, "learning_rate": 4.817298647376318e-06, "loss": 0.6095, "step": 230 }, { "epoch": 0.5416078984485191, "grad_norm": 0.6762199285625324, "learning_rate": 4.794551496292452e-06, "loss": 0.6082, "step": 240 }, { "epoch": 0.5641748942172073, "grad_norm": 0.5902874618389888, "learning_rate": 4.770537403418962e-06, "loss": 0.6022, "step": 250 }, { "epoch": 0.5867418899858956, "grad_norm": 0.7370791261766119, "learning_rate": 4.745271249504931e-06, "loss": 0.6047, "step": 260 }, { "epoch": 0.609308885754584, "grad_norm": 0.5946622411624806, "learning_rate": 4.7187686911607775e-06, "loss": 0.6, "step": 270 }, { "epoch": 0.6318758815232722, "grad_norm": 0.5496703727002791, "learning_rate": 4.691046151156365e-06, "loss": 0.6066, "step": 280 }, { "epoch": 0.6544428772919605, "grad_norm": 0.611792585555619, "learning_rate": 4.662120808244348e-06, "loss": 0.6, "step": 290 }, { "epoch": 0.6770098730606487, "grad_norm": 0.636806838675402, "learning_rate": 4.632010586515062e-06, "loss": 0.606, "step": 300 }, { "epoch": 0.6995768688293371, "grad_norm": 0.5554289199088291, "learning_rate": 4.60073414428955e-06, "loss": 0.599, "step": 310 }, { "epoch": 0.7221438645980254, "grad_norm": 0.5279748367952753, "learning_rate": 4.568310862557606e-06, "loss": 0.6021, "step": 320 }, { "epoch": 0.7447108603667136, "grad_norm": 0.5530959182218391, "learning_rate": 4.5347608329680125e-06, "loss": 0.5976, "step": 330 }, { "epoch": 0.767277856135402, "grad_norm": 0.5752251722608315, "learning_rate": 4.500104845378404e-06, "loss": 0.5969, "step": 340 }, { "epoch": 0.7898448519040903, "grad_norm": 0.5813755083202762, "learning_rate": 4.4643643749724655e-06, "loss": 0.5982, "step": 350 }, { "epoch": 0.8124118476727785, "grad_norm": 0.5973875786941857, "learning_rate": 4.427561568952466e-06, "loss": 0.5972, "step": 360 }, { "epoch": 0.8349788434414669, "grad_norm": 0.6033174529274596, "learning_rate": 4.389719232815361e-06, "loss": 0.6042, "step": 370 }, { "epoch": 0.8575458392101551, "grad_norm": 0.5901273574522466, "learning_rate": 4.350860816220966e-06, "loss": 0.6008, "step": 380 }, { "epoch": 0.8801128349788434, "grad_norm": 0.593608082407707, "learning_rate": 4.311010398460978e-06, "loss": 0.5947, "step": 390 }, { "epoch": 0.9026798307475318, "grad_norm": 0.5924882443981238, "learning_rate": 4.270192673537814e-06, "loss": 0.5968, "step": 400 }, { "epoch": 0.92524682651622, "grad_norm": 0.595780117352915, "learning_rate": 4.22843293486255e-06, "loss": 0.5948, "step": 410 }, { "epoch": 0.9478138222849083, "grad_norm": 0.6248231208615119, "learning_rate": 4.1857570595814195e-06, "loss": 0.5952, "step": 420 }, { "epoch": 0.9703808180535967, "grad_norm": 0.5148543249785058, "learning_rate": 4.142191492540592e-06, "loss": 0.5893, "step": 430 }, { "epoch": 0.9929478138222849, "grad_norm": 0.540989842605058, "learning_rate": 4.097763229899161e-06, "loss": 0.5939, "step": 440 }, { "epoch": 0.9997179125528914, "eval_loss": 0.5920091271400452, "eval_runtime": 680.7078, "eval_samples_per_second": 17.542, "eval_steps_per_second": 0.549, "step": 443 }, { "epoch": 1.0155148095909732, "grad_norm": 0.6626527903323465, "learning_rate": 4.052499802400512e-06, "loss": 0.6055, "step": 450 }, { "epoch": 1.0380818053596614, "grad_norm": 0.6475442970270617, "learning_rate": 4.006429258312416e-06, "loss": 0.5342, "step": 460 }, { "epoch": 1.0606488011283497, "grad_norm": 0.600016738279881, "learning_rate": 3.959580146046436e-06, "loss": 0.549, "step": 470 }, { "epoch": 1.0832157968970382, "grad_norm": 0.7191822331365615, "learning_rate": 3.9119814964674036e-06, "loss": 0.5365, "step": 480 }, { "epoch": 1.1057827926657264, "grad_norm": 0.6398896708187356, "learning_rate": 3.86366280490395e-06, "loss": 0.5469, "step": 490 }, { "epoch": 1.1283497884344147, "grad_norm": 0.555828522795445, "learning_rate": 3.814654012871206e-06, "loss": 0.5451, "step": 500 }, { "epoch": 1.150916784203103, "grad_norm": 0.5801058286608943, "learning_rate": 3.7649854895170252e-06, "loss": 0.5459, "step": 510 }, { "epoch": 1.1734837799717912, "grad_norm": 0.5685310953088956, "learning_rate": 3.7146880128032125e-06, "loss": 0.5383, "step": 520 }, { "epoch": 1.1960507757404795, "grad_norm": 0.5890904802760818, "learning_rate": 3.663792750433435e-06, "loss": 0.5439, "step": 530 }, { "epoch": 1.2186177715091677, "grad_norm": 0.6883190170025143, "learning_rate": 3.612331240539601e-06, "loss": 0.5413, "step": 540 }, { "epoch": 1.2411847672778562, "grad_norm": 0.67406923476649, "learning_rate": 3.5603353721387224e-06, "loss": 0.5462, "step": 550 }, { "epoch": 1.2637517630465445, "grad_norm": 0.5788655456772513, "learning_rate": 3.5078373653723222e-06, "loss": 0.5475, "step": 560 }, { "epoch": 1.2863187588152327, "grad_norm": 0.5830491880948598, "learning_rate": 3.454869751540672e-06, "loss": 0.5444, "step": 570 }, { "epoch": 1.308885754583921, "grad_norm": 0.5983182190975537, "learning_rate": 3.4014653529442025e-06, "loss": 0.5445, "step": 580 }, { "epoch": 1.3314527503526092, "grad_norm": 0.5628470529900238, "learning_rate": 3.3476572625446e-06, "loss": 0.5442, "step": 590 }, { "epoch": 1.3540197461212977, "grad_norm": 0.5502973875688251, "learning_rate": 3.2934788234581773e-06, "loss": 0.5494, "step": 600 }, { "epoch": 1.376586741889986, "grad_norm": 0.5801245589174187, "learning_rate": 3.238963608294232e-06, "loss": 0.5415, "step": 610 }, { "epoch": 1.3991537376586742, "grad_norm": 0.5821078689136515, "learning_rate": 3.1841453983511984e-06, "loss": 0.5459, "step": 620 }, { "epoch": 1.4217207334273625, "grad_norm": 0.5566752795206666, "learning_rate": 3.1290581626834814e-06, "loss": 0.5436, "step": 630 }, { "epoch": 1.4442877291960508, "grad_norm": 0.5598207245456932, "learning_rate": 3.0737360370519397e-06, "loss": 0.543, "step": 640 }, { "epoch": 1.466854724964739, "grad_norm": 0.5965882505937197, "learning_rate": 3.0182133027710727e-06, "loss": 0.5439, "step": 650 }, { "epoch": 1.4894217207334273, "grad_norm": 0.6050412455840206, "learning_rate": 2.9625243654660064e-06, "loss": 0.5471, "step": 660 }, { "epoch": 1.5119887165021155, "grad_norm": 0.5802695719407804, "learning_rate": 2.906703733752456e-06, "loss": 0.5419, "step": 670 }, { "epoch": 1.5345557122708038, "grad_norm": 0.5667065817171125, "learning_rate": 2.850785997852856e-06, "loss": 0.536, "step": 680 }, { "epoch": 1.5571227080394923, "grad_norm": 0.5693029540730941, "learning_rate": 2.7948058081619394e-06, "loss": 0.5377, "step": 690 }, { "epoch": 1.5796897038081805, "grad_norm": 0.5701466200858083, "learning_rate": 2.7387978537750094e-06, "loss": 0.5418, "step": 700 }, { "epoch": 1.6022566995768688, "grad_norm": 0.5422484900731249, "learning_rate": 2.6827968409922457e-06, "loss": 0.5444, "step": 710 }, { "epoch": 1.6248236953455573, "grad_norm": 0.5678359562664852, "learning_rate": 2.6268374718123436e-06, "loss": 0.5522, "step": 720 }, { "epoch": 1.6473906911142455, "grad_norm": 0.5651211493485119, "learning_rate": 2.5709544224288164e-06, "loss": 0.5354, "step": 730 }, { "epoch": 1.6699576868829338, "grad_norm": 0.5896266981049463, "learning_rate": 2.5151823217422937e-06, "loss": 0.5507, "step": 740 }, { "epoch": 1.692524682651622, "grad_norm": 0.561258191307828, "learning_rate": 2.4595557299021205e-06, "loss": 0.5422, "step": 750 }, { "epoch": 1.7150916784203103, "grad_norm": 0.5472432357761396, "learning_rate": 2.404109116890561e-06, "loss": 0.5409, "step": 760 }, { "epoch": 1.7376586741889986, "grad_norm": 0.5220118486994209, "learning_rate": 2.3488768411628787e-06, "loss": 0.5414, "step": 770 }, { "epoch": 1.7602256699576868, "grad_norm": 0.5643034542444902, "learning_rate": 2.29389312835652e-06, "loss": 0.5372, "step": 780 }, { "epoch": 1.782792665726375, "grad_norm": 0.5431143092086337, "learning_rate": 2.2391920500826093e-06, "loss": 0.5403, "step": 790 }, { "epoch": 1.8053596614950633, "grad_norm": 0.5551172023814825, "learning_rate": 2.184807502812879e-06, "loss": 0.5352, "step": 800 }, { "epoch": 1.8279266572637518, "grad_norm": 0.5221299415103324, "learning_rate": 2.1307731868751387e-06, "loss": 0.5458, "step": 810 }, { "epoch": 1.85049365303244, "grad_norm": 0.5224952013004204, "learning_rate": 2.0771225855702777e-06, "loss": 0.5329, "step": 820 }, { "epoch": 1.8730606488011283, "grad_norm": 0.5389042687272246, "learning_rate": 2.02388894442376e-06, "loss": 0.5447, "step": 830 }, { "epoch": 1.8956276445698168, "grad_norm": 0.53410794024608, "learning_rate": 1.9711052505844555e-06, "loss": 0.5449, "step": 840 }, { "epoch": 1.918194640338505, "grad_norm": 0.5366267684274532, "learning_rate": 1.918804212383584e-06, "loss": 0.5322, "step": 850 }, { "epoch": 1.9407616361071933, "grad_norm": 0.5700853439757274, "learning_rate": 1.867018239066429e-06, "loss": 0.5373, "step": 860 }, { "epoch": 1.9633286318758816, "grad_norm": 0.5364872137789395, "learning_rate": 1.815779420709386e-06, "loss": 0.5438, "step": 870 }, { "epoch": 1.9858956276445698, "grad_norm": 0.5302900665109113, "learning_rate": 1.7651195083347916e-06, "loss": 0.5415, "step": 880 }, { "epoch": 1.9994358251057829, "eval_loss": 0.5835571885108948, "eval_runtime": 685.3115, "eval_samples_per_second": 17.424, "eval_steps_per_second": 0.546, "step": 886 }, { "epoch": 2.008462623413258, "grad_norm": 0.7958964090979939, "learning_rate": 1.7150698942358537e-06, "loss": 0.5719, "step": 890 }, { "epoch": 2.0310296191819464, "grad_norm": 0.5965103763718388, "learning_rate": 1.6656615925238655e-06, "loss": 0.4955, "step": 900 }, { "epoch": 2.0535966149506346, "grad_norm": 0.5597696583743537, "learning_rate": 1.616925219909782e-06, "loss": 0.498, "step": 910 }, { "epoch": 2.076163610719323, "grad_norm": 0.5493791045621335, "learning_rate": 1.568890976732041e-06, "loss": 0.4901, "step": 920 }, { "epoch": 2.098730606488011, "grad_norm": 0.5448099821799507, "learning_rate": 1.5215886282423984e-06, "loss": 0.497, "step": 930 }, { "epoch": 2.1212976022566994, "grad_norm": 0.5621975505173159, "learning_rate": 1.4750474861613729e-06, "loss": 0.5006, "step": 940 }, { "epoch": 2.143864598025388, "grad_norm": 0.5559197950089757, "learning_rate": 1.4292963905147344e-06, "loss": 0.5016, "step": 950 }, { "epoch": 2.1664315937940763, "grad_norm": 0.5562082376298298, "learning_rate": 1.3843636917622792e-06, "loss": 0.4977, "step": 960 }, { "epoch": 2.1889985895627646, "grad_norm": 0.5648745989825581, "learning_rate": 1.3402772332299752e-06, "loss": 0.4945, "step": 970 }, { "epoch": 2.211565585331453, "grad_norm": 0.5445836211829971, "learning_rate": 1.2970643338563623e-06, "loss": 0.4889, "step": 980 }, { "epoch": 2.234132581100141, "grad_norm": 0.5404954635546201, "learning_rate": 1.2547517712639013e-06, "loss": 0.4886, "step": 990 }, { "epoch": 2.2566995768688294, "grad_norm": 0.560647006043489, "learning_rate": 1.2133657651657524e-06, "loss": 0.4978, "step": 1000 }, { "epoch": 2.2792665726375176, "grad_norm": 0.557442387654834, "learning_rate": 1.1729319611182772e-06, "loss": 0.5033, "step": 1010 }, { "epoch": 2.301833568406206, "grad_norm": 0.5683658622820689, "learning_rate": 1.1334754146293222e-06, "loss": 0.4972, "step": 1020 }, { "epoch": 2.324400564174894, "grad_norm": 0.590997174006606, "learning_rate": 1.095020575632144e-06, "loss": 0.5012, "step": 1030 }, { "epoch": 2.3469675599435824, "grad_norm": 0.5359140101612642, "learning_rate": 1.057591273334579e-06, "loss": 0.4985, "step": 1040 }, { "epoch": 2.3695345557122707, "grad_norm": 0.5483721423742447, "learning_rate": 1.0212107014528651e-06, "loss": 0.4958, "step": 1050 }, { "epoch": 2.392101551480959, "grad_norm": 0.5264325293497506, "learning_rate": 9.859014038392498e-07, "loss": 0.4949, "step": 1060 }, { "epoch": 2.414668547249647, "grad_norm": 0.5429027042593327, "learning_rate": 9.516852605123054e-07, "loss": 0.4951, "step": 1070 }, { "epoch": 2.4372355430183354, "grad_norm": 0.5876867058988207, "learning_rate": 9.185834740985922e-07, "loss": 0.4987, "step": 1080 }, { "epoch": 2.459802538787024, "grad_norm": 0.528249541513837, "learning_rate": 8.866165566940863e-07, "loss": 0.4967, "step": 1090 }, { "epoch": 2.4823695345557124, "grad_norm": 0.5411121695065451, "learning_rate": 8.55804317153501e-07, "loss": 0.4992, "step": 1100 }, { "epoch": 2.5049365303244007, "grad_norm": 0.5351045285030581, "learning_rate": 8.26165848815389e-07, "loss": 0.5022, "step": 1110 }, { "epoch": 2.527503526093089, "grad_norm": 0.5328584232809477, "learning_rate": 7.977195176706218e-07, "loss": 0.4921, "step": 1120 }, { "epoch": 2.550070521861777, "grad_norm": 0.593905343882913, "learning_rate": 7.704829509815869e-07, "loss": 0.497, "step": 1130 }, { "epoch": 2.5726375176304654, "grad_norm": 0.548485481272847, "learning_rate": 7.444730263591466e-07, "loss": 0.4951, "step": 1140 }, { "epoch": 2.5952045133991537, "grad_norm": 0.5460617858788939, "learning_rate": 7.197058613041364e-07, "loss": 0.4955, "step": 1150 }, { "epoch": 2.617771509167842, "grad_norm": 0.5749118090310457, "learning_rate": 6.961968032198748e-07, "loss": 0.4989, "step": 1160 }, { "epoch": 2.64033850493653, "grad_norm": 0.5283077843563921, "learning_rate": 6.739604199018791e-07, "loss": 0.4875, "step": 1170 }, { "epoch": 2.6629055007052185, "grad_norm": 0.5842336257146924, "learning_rate": 6.530104905106762e-07, "loss": 0.4985, "step": 1180 }, { "epoch": 2.685472496473907, "grad_norm": 0.551245465232118, "learning_rate": 6.333599970333109e-07, "loss": 0.4991, "step": 1190 }, { "epoch": 2.7080394922425954, "grad_norm": 0.5352944796561586, "learning_rate": 6.150211162388293e-07, "loss": 0.4974, "step": 1200 }, { "epoch": 2.7306064880112837, "grad_norm": 0.5389072174031843, "learning_rate": 5.980052121327338e-07, "loss": 0.498, "step": 1210 }, { "epoch": 2.753173483779972, "grad_norm": 0.5261647300698946, "learning_rate": 5.823228289150787e-07, "loss": 0.4945, "step": 1220 }, { "epoch": 2.77574047954866, "grad_norm": 0.5328094822460281, "learning_rate": 5.679836844465767e-07, "loss": 0.5013, "step": 1230 }, { "epoch": 2.7983074753173485, "grad_norm": 0.5267049638383339, "learning_rate": 5.549966642267552e-07, "loss": 0.4996, "step": 1240 }, { "epoch": 2.8208744710860367, "grad_norm": 0.5310439592442067, "learning_rate": 5.433698158879043e-07, "loss": 0.4992, "step": 1250 }, { "epoch": 2.843441466854725, "grad_norm": 0.5421196499338565, "learning_rate": 5.331103442082198e-07, "loss": 0.5009, "step": 1260 }, { "epoch": 2.8660084626234132, "grad_norm": 0.528752193816906, "learning_rate": 5.242246066472395e-07, "loss": 0.498, "step": 1270 }, { "epoch": 2.8885754583921015, "grad_norm": 0.5210158884775411, "learning_rate": 5.167181094063323e-07, "loss": 0.4955, "step": 1280 }, { "epoch": 2.9111424541607898, "grad_norm": 0.5435369205100163, "learning_rate": 5.105955040166854e-07, "loss": 0.4994, "step": 1290 }, { "epoch": 2.933709449929478, "grad_norm": 0.5330835467051688, "learning_rate": 5.058605844569015e-07, "loss": 0.4945, "step": 1300 }, { "epoch": 2.9562764456981663, "grad_norm": 0.548558970817452, "learning_rate": 5.025162848019948e-07, "loss": 0.4985, "step": 1310 }, { "epoch": 2.9788434414668545, "grad_norm": 0.5379965054372117, "learning_rate": 5.00564677405241e-07, "loss": 0.4941, "step": 1320 }, { "epoch": 2.9991537376586743, "eval_loss": 0.5908229947090149, "eval_runtime": 682.7955, "eval_samples_per_second": 17.488, "eval_steps_per_second": 0.548, "step": 1329 }, { "epoch": 2.9991537376586743, "step": 1329, "total_flos": 5064195066298368.0, "train_loss": 0.554891100855498, "train_runtime": 119717.367, "train_samples_per_second": 5.685, "train_steps_per_second": 0.011 } ], "logging_steps": 10, "max_steps": 1329, "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": 5064195066298368.0, "train_batch_size": 4, "trial_name": null, "trial_params": null }