| { | |
| "best_global_step": null, | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 5.0, | |
| "eval_steps": 500, | |
| "global_step": 2075, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.060350030175015085, | |
| "grad_norm": 0.22048431634902954, | |
| "learning_rate": 8.18181818181818e-05, | |
| "loss": 1.8437, | |
| "mean_token_accuracy": 0.6185667091608047, | |
| "num_tokens": 157017.0, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.12070006035003017, | |
| "grad_norm": 0.28467562794685364, | |
| "learning_rate": 0.00016704545454545452, | |
| "loss": 1.0769, | |
| "mean_token_accuracy": 0.7381608086824417, | |
| "num_tokens": 282861.0, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.18105009052504525, | |
| "grad_norm": 0.27485042810440063, | |
| "learning_rate": 0.0002522727272727273, | |
| "loss": 0.6762, | |
| "mean_token_accuracy": 0.8168852615356446, | |
| "num_tokens": 439048.0, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.24140012070006034, | |
| "grad_norm": 0.3554552495479584, | |
| "learning_rate": 0.0002999887132933212, | |
| "loss": 0.5642, | |
| "mean_token_accuracy": 0.8437625843286515, | |
| "num_tokens": 564700.0, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.30175015087507545, | |
| "grad_norm": 0.24375054240226746, | |
| "learning_rate": 0.0002998791256978121, | |
| "loss": 0.4572, | |
| "mean_token_accuracy": 0.869450313448906, | |
| "num_tokens": 720573.0, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.3621001810500905, | |
| "grad_norm": 0.33107709884643555, | |
| "learning_rate": 0.0002996530399366737, | |
| "loss": 0.3565, | |
| "mean_token_accuracy": 0.8982734948396682, | |
| "num_tokens": 847195.0, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.4224502112251056, | |
| "grad_norm": 0.30305179953575134, | |
| "learning_rate": 0.00029931063174202567, | |
| "loss": 0.2814, | |
| "mean_token_accuracy": 0.9188993269205094, | |
| "num_tokens": 1005729.0, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.4828002414001207, | |
| "grad_norm": 0.35704073309898376, | |
| "learning_rate": 0.00029885216726118104, | |
| "loss": 0.2394, | |
| "mean_token_accuracy": 0.9293693268299102, | |
| "num_tokens": 1132299.0, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.5431502715751357, | |
| "grad_norm": 0.26589134335517883, | |
| "learning_rate": 0.00029827800284977474, | |
| "loss": 0.1821, | |
| "mean_token_accuracy": 0.9475246322154999, | |
| "num_tokens": 1291425.0, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 0.6035003017501509, | |
| "grad_norm": 0.3292100131511688, | |
| "learning_rate": 0.00029758858479477575, | |
| "loss": 0.1871, | |
| "mean_token_accuracy": 0.9475667542219162, | |
| "num_tokens": 1417320.0, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.663850331925166, | |
| "grad_norm": 0.30751854181289673, | |
| "learning_rate": 0.0002967844489675963, | |
| "loss": 0.1468, | |
| "mean_token_accuracy": 0.9586230200529099, | |
| "num_tokens": 1575826.0, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 0.724200362100181, | |
| "grad_norm": 0.28926485776901245, | |
| "learning_rate": 0.00029586622040756957, | |
| "loss": 0.1372, | |
| "mean_token_accuracy": 0.9609356409311295, | |
| "num_tokens": 1702190.0, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.7845503922751962, | |
| "grad_norm": 0.2157151848077774, | |
| "learning_rate": 0.0002948346128361186, | |
| "loss": 0.1159, | |
| "mean_token_accuracy": 0.9670810961723327, | |
| "num_tokens": 1860388.0, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 0.8449004224502112, | |
| "grad_norm": 0.3129127621650696, | |
| "learning_rate": 0.00029369042810199416, | |
| "loss": 0.0997, | |
| "mean_token_accuracy": 0.9725474363565445, | |
| "num_tokens": 1988000.0, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.9052504526252263, | |
| "grad_norm": 0.12445805966854095, | |
| "learning_rate": 0.0002924345555580135, | |
| "loss": 0.099, | |
| "mean_token_accuracy": 0.9723707520961762, | |
| "num_tokens": 2146759.0, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 0.9656004828002414, | |
| "grad_norm": 0.2350274920463562, | |
| "learning_rate": 0.000291067971369783, | |
| "loss": 0.087, | |
| "mean_token_accuracy": 0.9752390831708908, | |
| "num_tokens": 2274851.0, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.08540041744709015, | |
| "eval_mean_token_accuracy": 0.9763027584230578, | |
| "eval_num_tokens": 2354180.0, | |
| "eval_runtime": 62.666, | |
| "eval_samples_per_second": 5.888, | |
| "eval_steps_per_second": 2.952, | |
| "step": 415 | |
| }, | |
| { | |
| "epoch": 1.024140012070006, | |
| "grad_norm": 0.1618579775094986, | |
| "learning_rate": 0.0002895917377569438, | |
| "loss": 0.0946, | |
| "mean_token_accuracy": 0.9745458819202542, | |
| "num_tokens": 2424550.0, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 1.0844900422450212, | |
| "grad_norm": 0.17064224183559418, | |
| "learning_rate": 0.00028800700216752875, | |
| "loss": 0.059, | |
| "mean_token_accuracy": 0.9832033175230026, | |
| "num_tokens": 2565889.0, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 1.1448400724200363, | |
| "grad_norm": 0.15575812757015228, | |
| "learning_rate": 0.00028631499638607285, | |
| "loss": 0.0719, | |
| "mean_token_accuracy": 0.9796904844045639, | |
| "num_tokens": 2707834.0, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 1.2051901025950513, | |
| "grad_norm": 0.15536518394947052, | |
| "learning_rate": 0.0002845170355761712, | |
| "loss": 0.0586, | |
| "mean_token_accuracy": 0.9835732167959214, | |
| "num_tokens": 2850768.0, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.2655401327700664, | |
| "grad_norm": 0.13807828724384308, | |
| "learning_rate": 0.0002826145172582274, | |
| "loss": 0.0645, | |
| "mean_token_accuracy": 0.9819038778543472, | |
| "num_tokens": 2990877.0, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 1.3258901629450814, | |
| "grad_norm": 0.2444400042295456, | |
| "learning_rate": 0.00028060892022318764, | |
| "loss": 0.049, | |
| "mean_token_accuracy": 0.9859754049777985, | |
| "num_tokens": 3131813.0, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 1.3862401931200965, | |
| "grad_norm": 0.16260553896427155, | |
| "learning_rate": 0.0002785018033831051, | |
| "loss": 0.0724, | |
| "mean_token_accuracy": 0.9795020925998688, | |
| "num_tokens": 3271944.0, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 1.4465902232951118, | |
| "grad_norm": 0.15038788318634033, | |
| "learning_rate": 0.0002762948045594276, | |
| "loss": 0.048, | |
| "mean_token_accuracy": 0.986338136792183, | |
| "num_tokens": 3415529.0, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.5069402534701268, | |
| "grad_norm": 0.12307009845972061, | |
| "learning_rate": 0.0002739896392099502, | |
| "loss": 0.0648, | |
| "mean_token_accuracy": 0.9821051049232483, | |
| "num_tokens": 3557697.0, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 1.567290283645142, | |
| "grad_norm": 0.17994529008865356, | |
| "learning_rate": 0.00027158809909542307, | |
| "loss": 0.0458, | |
| "mean_token_accuracy": 0.9869296491146088, | |
| "num_tokens": 3701711.0, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 1.627640313820157, | |
| "grad_norm": 0.12831450998783112, | |
| "learning_rate": 0.00026909205088685, | |
| "loss": 0.0602, | |
| "mean_token_accuracy": 0.9828966480493545, | |
| "num_tokens": 3844440.0, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 1.687990343995172, | |
| "grad_norm": 0.12778432667255402, | |
| "learning_rate": 0.0002665034347145612, | |
| "loss": 0.0492, | |
| "mean_token_accuracy": 0.9861533933877945, | |
| "num_tokens": 3986855.0, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.748340374170187, | |
| "grad_norm": 0.09499543905258179, | |
| "learning_rate": 0.000263824262660187, | |
| "loss": 0.0561, | |
| "mean_token_accuracy": 0.984484378695488, | |
| "num_tokens": 4128398.0, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 1.8086904043452021, | |
| "grad_norm": 0.15675826370716095, | |
| "learning_rate": 0.0002610566171927056, | |
| "loss": 0.0426, | |
| "mean_token_accuracy": 0.9876346814632416, | |
| "num_tokens": 4270406.0, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 1.8690404345202172, | |
| "grad_norm": 0.11688172817230225, | |
| "learning_rate": 0.00025820264954977976, | |
| "loss": 0.0565, | |
| "mean_token_accuracy": 0.9839567536115646, | |
| "num_tokens": 4412708.0, | |
| "step": 775 | |
| }, | |
| { | |
| "epoch": 1.9293904646952322, | |
| "grad_norm": 0.1486789733171463, | |
| "learning_rate": 0.00025526457806564136, | |
| "loss": 0.0417, | |
| "mean_token_accuracy": 0.9880886596441268, | |
| "num_tokens": 4556150.0, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.9897404948702473, | |
| "grad_norm": 0.11592721939086914, | |
| "learning_rate": 0.00025224468644682245, | |
| "loss": 0.0492, | |
| "mean_token_accuracy": 0.9864854019880295, | |
| "num_tokens": 4688551.0, | |
| "step": 825 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.054148148745298386, | |
| "eval_mean_token_accuracy": 0.9855220913887024, | |
| "eval_num_tokens": 4708360.0, | |
| "eval_runtime": 62.7544, | |
| "eval_samples_per_second": 5.88, | |
| "eval_steps_per_second": 2.948, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 2.048280024140012, | |
| "grad_norm": 0.1103988066315651, | |
| "learning_rate": 0.00024914532199707444, | |
| "loss": 0.0456, | |
| "mean_token_accuracy": 0.9862007431148254, | |
| "num_tokens": 4838545.0, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 2.1086300543150274, | |
| "grad_norm": 0.17865443229675293, | |
| "learning_rate": 0.00024596889379285353, | |
| "loss": 0.0335, | |
| "mean_token_accuracy": 0.9900023031234741, | |
| "num_tokens": 4971291.0, | |
| "step": 875 | |
| }, | |
| { | |
| "epoch": 2.1689800844900424, | |
| "grad_norm": 0.08458594232797623, | |
| "learning_rate": 0.00024271787081079228, | |
| "loss": 0.0445, | |
| "mean_token_accuracy": 0.9871802502870559, | |
| "num_tokens": 5122538.0, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 2.2293301146650575, | |
| "grad_norm": 0.0892493948340416, | |
| "learning_rate": 0.00023939478000861117, | |
| "loss": 0.0347, | |
| "mean_token_accuracy": 0.9895845329761506, | |
| "num_tokens": 5256508.0, | |
| "step": 925 | |
| }, | |
| { | |
| "epoch": 2.2896801448400725, | |
| "grad_norm": 0.08608334511518478, | |
| "learning_rate": 0.00023600220436096318, | |
| "loss": 0.0428, | |
| "mean_token_accuracy": 0.9870429944992065, | |
| "num_tokens": 5406644.0, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 2.3500301750150876, | |
| "grad_norm": 0.11120013892650604, | |
| "learning_rate": 0.00023254278085173684, | |
| "loss": 0.0304, | |
| "mean_token_accuracy": 0.990639505982399, | |
| "num_tokens": 5539678.0, | |
| "step": 975 | |
| }, | |
| { | |
| "epoch": 2.4103802051901027, | |
| "grad_norm": 0.08932222425937653, | |
| "learning_rate": 0.00022901919842437972, | |
| "loss": 0.0453, | |
| "mean_token_accuracy": 0.9866666996479034, | |
| "num_tokens": 5692025.0, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.4707302353651177, | |
| "grad_norm": 0.07231656461954117, | |
| "learning_rate": 0.00022543419589183397, | |
| "loss": 0.031, | |
| "mean_token_accuracy": 0.990651159286499, | |
| "num_tokens": 5824644.0, | |
| "step": 1025 | |
| }, | |
| { | |
| "epoch": 2.5310802655401328, | |
| "grad_norm": 0.09120254963636398, | |
| "learning_rate": 0.00022179055980770993, | |
| "loss": 0.0421, | |
| "mean_token_accuracy": 0.9870923537015915, | |
| "num_tokens": 5974858.0, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 2.591430295715148, | |
| "grad_norm": 0.09927839785814285, | |
| "learning_rate": 0.0002180911223003513, | |
| "loss": 0.0294, | |
| "mean_token_accuracy": 0.9910473513603211, | |
| "num_tokens": 6106888.0, | |
| "step": 1075 | |
| }, | |
| { | |
| "epoch": 2.651780325890163, | |
| "grad_norm": 0.1529274433851242, | |
| "learning_rate": 0.00021433875887147627, | |
| "loss": 0.0398, | |
| "mean_token_accuracy": 0.9881814271211624, | |
| "num_tokens": 6257087.0, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 2.712130356065178, | |
| "grad_norm": 0.07802440971136093, | |
| "learning_rate": 0.00021053638616110525, | |
| "loss": 0.0313, | |
| "mean_token_accuracy": 0.9907257658243179, | |
| "num_tokens": 6390499.0, | |
| "step": 1125 | |
| }, | |
| { | |
| "epoch": 2.772480386240193, | |
| "grad_norm": 0.06188732013106346, | |
| "learning_rate": 0.00020668695968051274, | |
| "loss": 0.0463, | |
| "mean_token_accuracy": 0.9869219189882279, | |
| "num_tokens": 6542392.0, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 2.832830416415208, | |
| "grad_norm": 0.09649744629859924, | |
| "learning_rate": 0.00020279347151496482, | |
| "loss": 0.0278, | |
| "mean_token_accuracy": 0.9914505350589752, | |
| "num_tokens": 6675191.0, | |
| "step": 1175 | |
| }, | |
| { | |
| "epoch": 2.8931804465902236, | |
| "grad_norm": 0.05005449801683426, | |
| "learning_rate": 0.00019885894799802922, | |
| "loss": 0.0452, | |
| "mean_token_accuracy": 0.9864512634277344, | |
| "num_tokens": 6828887.0, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 2.9535304767652386, | |
| "grad_norm": 0.09346788376569748, | |
| "learning_rate": 0.00019488644735926396, | |
| "loss": 0.0317, | |
| "mean_token_accuracy": 0.9907874202728272, | |
| "num_tokens": 6962022.0, | |
| "step": 1225 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 0.04395000636577606, | |
| "eval_mean_token_accuracy": 0.9879707652169305, | |
| "eval_num_tokens": 7062540.0, | |
| "eval_runtime": 62.7009, | |
| "eval_samples_per_second": 5.885, | |
| "eval_steps_per_second": 2.951, | |
| "step": 1245 | |
| }, | |
| { | |
| "epoch": 3.012070006035003, | |
| "grad_norm": 0.10707546025514603, | |
| "learning_rate": 0.00019087905734711452, | |
| "loss": 0.0388, | |
| "mean_token_accuracy": 0.9883427564630803, | |
| "num_tokens": 7099771.0, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 3.0724200362100182, | |
| "grad_norm": 0.04371380805969238, | |
| "learning_rate": 0.00018683989282886613, | |
| "loss": 0.0259, | |
| "mean_token_accuracy": 0.9917033529281616, | |
| "num_tokens": 7247681.0, | |
| "step": 1275 | |
| }, | |
| { | |
| "epoch": 3.1327700663850333, | |
| "grad_norm": 0.0800153911113739, | |
| "learning_rate": 0.0001827720933695173, | |
| "loss": 0.0308, | |
| "mean_token_accuracy": 0.9905261534452439, | |
| "num_tokens": 7381955.0, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 3.1931200965600484, | |
| "grad_norm": 0.09413493424654007, | |
| "learning_rate": 0.00017867882079145627, | |
| "loss": 0.0263, | |
| "mean_token_accuracy": 0.9920635151863099, | |
| "num_tokens": 7528804.0, | |
| "step": 1325 | |
| }, | |
| { | |
| "epoch": 3.2534701267350634, | |
| "grad_norm": 0.07196678966283798, | |
| "learning_rate": 0.00017456325671683724, | |
| "loss": 0.0319, | |
| "mean_token_accuracy": 0.9904261428117752, | |
| "num_tokens": 7664857.0, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 3.3138201569100785, | |
| "grad_norm": 0.07312282174825668, | |
| "learning_rate": 0.00017042860009456638, | |
| "loss": 0.0267, | |
| "mean_token_accuracy": 0.9914035642147064, | |
| "num_tokens": 7812961.0, | |
| "step": 1375 | |
| }, | |
| { | |
| "epoch": 3.3741701870850935, | |
| "grad_norm": 0.07159588485956192, | |
| "learning_rate": 0.00016627806471382066, | |
| "loss": 0.0282, | |
| "mean_token_accuracy": 0.9908248996734619, | |
| "num_tokens": 7948932.0, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 3.4345202172601086, | |
| "grad_norm": 0.04767977446317673, | |
| "learning_rate": 0.00016211487670603078, | |
| "loss": 0.0259, | |
| "mean_token_accuracy": 0.9916666203737259, | |
| "num_tokens": 8097306.0, | |
| "step": 1425 | |
| }, | |
| { | |
| "epoch": 3.4948702474351236, | |
| "grad_norm": 0.061906732618808746, | |
| "learning_rate": 0.0001579422720372715, | |
| "loss": 0.0269, | |
| "mean_token_accuracy": 0.9917305141687394, | |
| "num_tokens": 8232788.0, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 3.5552202776101387, | |
| "grad_norm": 0.04637061804533005, | |
| "learning_rate": 0.00015376349399300745, | |
| "loss": 0.0263, | |
| "mean_token_accuracy": 0.9916825985908508, | |
| "num_tokens": 8381745.0, | |
| "step": 1475 | |
| }, | |
| { | |
| "epoch": 3.6155703077851538, | |
| "grad_norm": 0.12033607810735703, | |
| "learning_rate": 0.0001495817906571492, | |
| "loss": 0.0322, | |
| "mean_token_accuracy": 0.9897854518890381, | |
| "num_tokens": 8519204.0, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 3.675920337960169, | |
| "grad_norm": 0.06281058490276337, | |
| "learning_rate": 0.00014540041238738055, | |
| "loss": 0.0249, | |
| "mean_token_accuracy": 0.9918615108728409, | |
| "num_tokens": 8666692.0, | |
| "step": 1525 | |
| }, | |
| { | |
| "epoch": 3.736270368135184, | |
| "grad_norm": 0.06974755227565765, | |
| "learning_rate": 0.00014122260928871734, | |
| "loss": 0.0299, | |
| "mean_token_accuracy": 0.9906923991441726, | |
| "num_tokens": 8802593.0, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 3.796620398310199, | |
| "grad_norm": 0.048978209495544434, | |
| "learning_rate": 0.00013705162868726396, | |
| "loss": 0.0258, | |
| "mean_token_accuracy": 0.9917961621284485, | |
| "num_tokens": 8951230.0, | |
| "step": 1575 | |
| }, | |
| { | |
| "epoch": 3.856970428485214, | |
| "grad_norm": 0.0680331289768219, | |
| "learning_rate": 0.00013289071260612855, | |
| "loss": 0.0269, | |
| "mean_token_accuracy": 0.991482435464859, | |
| "num_tokens": 9086196.0, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 3.9173204586602295, | |
| "grad_norm": 0.04798055812716484, | |
| "learning_rate": 0.00012874309524546083, | |
| "loss": 0.025, | |
| "mean_token_accuracy": 0.9918987530469895, | |
| "num_tokens": 9235448.0, | |
| "step": 1625 | |
| }, | |
| { | |
| "epoch": 3.9776704888352445, | |
| "grad_norm": 0.07034997642040253, | |
| "learning_rate": 0.00012461200046857084, | |
| "loss": 0.0244, | |
| "mean_token_accuracy": 0.9922022837400436, | |
| "num_tokens": 9368109.0, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 0.04039536416530609, | |
| "eval_mean_token_accuracy": 0.9890036863249702, | |
| "eval_num_tokens": 9416720.0, | |
| "eval_runtime": 62.709, | |
| "eval_samples_per_second": 5.884, | |
| "eval_steps_per_second": 2.95, | |
| "step": 1660 | |
| }, | |
| { | |
| "epoch": 4.036210018105009, | |
| "grad_norm": 0.04823560640215874, | |
| "learning_rate": 0.00012050063929608123, | |
| "loss": 0.0245, | |
| "mean_token_accuracy": 0.9920303588060989, | |
| "num_tokens": 9515827.0, | |
| "step": 1675 | |
| }, | |
| { | |
| "epoch": 4.096560048280024, | |
| "grad_norm": 0.04038386419415474, | |
| "learning_rate": 0.0001164122074100633, | |
| "loss": 0.0205, | |
| "mean_token_accuracy": 0.9934929460287094, | |
| "num_tokens": 9653322.0, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 4.15691007845504, | |
| "grad_norm": 0.07340297102928162, | |
| "learning_rate": 0.00011234988267009415, | |
| "loss": 0.024, | |
| "mean_token_accuracy": 0.9922641849517823, | |
| "num_tokens": 9799350.0, | |
| "step": 1725 | |
| }, | |
| { | |
| "epoch": 4.217260108630055, | |
| "grad_norm": 0.038047417998313904, | |
| "learning_rate": 0.00010831682264316787, | |
| "loss": 0.0199, | |
| "mean_token_accuracy": 0.9933889234066009, | |
| "num_tokens": 9936097.0, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 4.27761013880507, | |
| "grad_norm": 0.13129326701164246, | |
| "learning_rate": 0.00010431616214937911, | |
| "loss": 0.0233, | |
| "mean_token_accuracy": 0.9926201468706131, | |
| "num_tokens": 10082564.0, | |
| "step": 1775 | |
| }, | |
| { | |
| "epoch": 4.337960168980085, | |
| "grad_norm": 0.037602320313453674, | |
| "learning_rate": 0.00010035101082528777, | |
| "loss": 0.0205, | |
| "mean_token_accuracy": 0.9932494872808456, | |
| "num_tokens": 10219755.0, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 4.3983101991551, | |
| "grad_norm": 0.04699968174099922, | |
| "learning_rate": 9.642445070685809e-05, | |
| "loss": 0.0245, | |
| "mean_token_accuracy": 0.992087431550026, | |
| "num_tokens": 10367027.0, | |
| "step": 1825 | |
| }, | |
| { | |
| "epoch": 4.458660229330115, | |
| "grad_norm": 0.04391399025917053, | |
| "learning_rate": 9.253953383385157e-05, | |
| "loss": 0.0196, | |
| "mean_token_accuracy": 0.9935489571094513, | |
| "num_tokens": 10506017.0, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 4.51901025950513, | |
| "grad_norm": 0.04784134775400162, | |
| "learning_rate": 8.869927987753459e-05, | |
| "loss": 0.0232, | |
| "mean_token_accuracy": 0.9924498957395553, | |
| "num_tokens": 10652825.0, | |
| "step": 1875 | |
| }, | |
| { | |
| "epoch": 4.579360289680145, | |
| "grad_norm": 0.04309968650341034, | |
| "learning_rate": 8.490667379354661e-05, | |
| "loss": 0.0189, | |
| "mean_token_accuracy": 0.9937076556682587, | |
| "num_tokens": 10790476.0, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 4.63971031985516, | |
| "grad_norm": 0.07193579524755478, | |
| "learning_rate": 8.116466350175079e-05, | |
| "loss": 0.0235, | |
| "mean_token_accuracy": 0.992224183678627, | |
| "num_tokens": 10935736.0, | |
| "step": 1925 | |
| }, | |
| { | |
| "epoch": 4.700060350030175, | |
| "grad_norm": 0.04449706897139549, | |
| "learning_rate": 7.747615759487304e-05, | |
| "loss": 0.0194, | |
| "mean_token_accuracy": 0.9936429464817047, | |
| "num_tokens": 11072468.0, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 4.76041038020519, | |
| "grad_norm": 0.06068187579512596, | |
| "learning_rate": 7.38440230777085e-05, | |
| "loss": 0.0235, | |
| "mean_token_accuracy": 0.992416363954544, | |
| "num_tokens": 11218509.0, | |
| "step": 1975 | |
| }, | |
| { | |
| "epoch": 4.820760410380205, | |
| "grad_norm": 0.030500777065753937, | |
| "learning_rate": 7.027108313865378e-05, | |
| "loss": 0.0185, | |
| "mean_token_accuracy": 0.9937655937671661, | |
| "num_tokens": 11356841.0, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 4.88111044055522, | |
| "grad_norm": 0.1406712532043457, | |
| "learning_rate": 6.676011495529687e-05, | |
| "loss": 0.0243, | |
| "mean_token_accuracy": 0.9923334455490113, | |
| "num_tokens": 11505617.0, | |
| "step": 2025 | |
| }, | |
| { | |
| "epoch": 4.941460470730235, | |
| "grad_norm": 0.0490015484392643, | |
| "learning_rate": 6.331384753577056e-05, | |
| "loss": 0.0188, | |
| "mean_token_accuracy": 0.993909627199173, | |
| "num_tokens": 11642628.0, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 0.13373523950576782, | |
| "learning_rate": 5.993495959754631e-05, | |
| "loss": 0.0221, | |
| "mean_token_accuracy": 0.992490774577426, | |
| "num_tokens": 11770900.0, | |
| "step": 2075 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 0.03932187706232071, | |
| "eval_mean_token_accuracy": 0.9896563478418299, | |
| "eval_num_tokens": 11770900.0, | |
| "eval_runtime": 62.7147, | |
| "eval_samples_per_second": 5.884, | |
| "eval_steps_per_second": 2.95, | |
| "step": 2075 | |
| } | |
| ], | |
| "logging_steps": 25, | |
| "max_steps": 2905, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 7, | |
| "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": 5.0566765205027635e+17, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |