| { | |
| "best_global_step": null, | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.983132530120482, | |
| "eval_steps": 20, | |
| "global_step": 204, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.03855421686746988, | |
| "grad_norm": 8.12320851233868, | |
| "learning_rate": 9.999407114490384e-05, | |
| "loss": 0.3959, | |
| "mean_token_accuracy": 0.9143258593976498, | |
| "num_tokens": 131072.0, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.07710843373493977, | |
| "grad_norm": 3.806704898812171, | |
| "learning_rate": 9.994664874011863e-05, | |
| "loss": 0.4698, | |
| "mean_token_accuracy": 0.8628251552581787, | |
| "num_tokens": 262144.0, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.11566265060240964, | |
| "grad_norm": 1.3442597047314044, | |
| "learning_rate": 9.985184891357164e-05, | |
| "loss": 0.3588, | |
| "mean_token_accuracy": 0.889327384531498, | |
| "num_tokens": 393216.0, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.15421686746987953, | |
| "grad_norm": 0.9380641398553072, | |
| "learning_rate": 9.970976158864073e-05, | |
| "loss": 0.3096, | |
| "mean_token_accuracy": 0.9009984359145164, | |
| "num_tokens": 524288.0, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.1927710843373494, | |
| "grad_norm": 0.8735334897676842, | |
| "learning_rate": 9.952052154376026e-05, | |
| "loss": 0.309, | |
| "mean_token_accuracy": 0.9008094593882561, | |
| "num_tokens": 654484.0, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.23132530120481928, | |
| "grad_norm": 0.7148108689183823, | |
| "learning_rate": 9.928430828457572e-05, | |
| "loss": 0.3241, | |
| "mean_token_accuracy": 0.8963193334639072, | |
| "num_tokens": 785556.0, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.26987951807228916, | |
| "grad_norm": 0.6963169490883665, | |
| "learning_rate": 9.90013458736716e-05, | |
| "loss": 0.3229, | |
| "mean_token_accuracy": 0.8985304310917854, | |
| "num_tokens": 915519.0, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.30843373493975906, | |
| "grad_norm": 0.7184542724122979, | |
| "learning_rate": 9.867190271803465e-05, | |
| "loss": 0.3174, | |
| "mean_token_accuracy": 0.898639801889658, | |
| "num_tokens": 1046591.0, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.3469879518072289, | |
| "grad_norm": 0.7018308439433193, | |
| "learning_rate": 9.829629131445342e-05, | |
| "loss": 0.322, | |
| "mean_token_accuracy": 0.8966093920171261, | |
| "num_tokens": 1177663.0, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.3855421686746988, | |
| "grad_norm": 0.6150629034715422, | |
| "learning_rate": 9.787486795309621e-05, | |
| "loss": 0.3226, | |
| "mean_token_accuracy": 0.8969528824090958, | |
| "num_tokens": 1308735.0, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.3855421686746988, | |
| "eval_loss": 0.8887339234352112, | |
| "eval_mean_token_accuracy": 0.7687314558029175, | |
| "eval_num_tokens": 1308735.0, | |
| "eval_runtime": 15.4789, | |
| "eval_samples_per_second": 25.583, | |
| "eval_steps_per_second": 3.23, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.42409638554216866, | |
| "grad_norm": 0.6188818540117794, | |
| "learning_rate": 9.74080323795483e-05, | |
| "loss": 0.3111, | |
| "mean_token_accuracy": 0.8995252437889576, | |
| "num_tokens": 1439807.0, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.46265060240963857, | |
| "grad_norm": 6.76038795199159, | |
| "learning_rate": 9.689622741562892e-05, | |
| "loss": 0.3566, | |
| "mean_token_accuracy": 0.8917263597249985, | |
| "num_tokens": 1570062.0, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.5012048192771085, | |
| "grad_norm": 0.7901190162268289, | |
| "learning_rate": 9.633993853934803e-05, | |
| "loss": 0.3315, | |
| "mean_token_accuracy": 0.8946247734129429, | |
| "num_tokens": 1701134.0, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.5397590361445783, | |
| "grad_norm": 0.6533302263211975, | |
| "learning_rate": 9.573969342440106e-05, | |
| "loss": 0.3223, | |
| "mean_token_accuracy": 0.8971585147082806, | |
| "num_tokens": 1832133.0, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.5783132530120482, | |
| "grad_norm": 0.5629583530778991, | |
| "learning_rate": 9.509606143963832e-05, | |
| "loss": 0.3296, | |
| "mean_token_accuracy": 0.8950598686933517, | |
| "num_tokens": 1963205.0, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.6168674698795181, | |
| "grad_norm": 0.5693916351191454, | |
| "learning_rate": 9.440965310898424e-05, | |
| "loss": 0.3192, | |
| "mean_token_accuracy": 0.8991741202771664, | |
| "num_tokens": 2094277.0, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.655421686746988, | |
| "grad_norm": 0.5746142310558011, | |
| "learning_rate": 9.368111953231848e-05, | |
| "loss": 0.3172, | |
| "mean_token_accuracy": 0.8990138210356236, | |
| "num_tokens": 2225349.0, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.6939759036144578, | |
| "grad_norm": 0.5900722105723064, | |
| "learning_rate": 9.291115176786814e-05, | |
| "loss": 0.3088, | |
| "mean_token_accuracy": 0.9014212191104889, | |
| "num_tokens": 2355263.0, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.7325301204819277, | |
| "grad_norm": 16.33773536913975, | |
| "learning_rate": 9.210048017669726e-05, | |
| "loss": 0.3646, | |
| "mean_token_accuracy": 0.8937164358794689, | |
| "num_tokens": 2486335.0, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.7710843373493976, | |
| "grad_norm": 0.6405118103192932, | |
| "learning_rate": 9.124987372991511e-05, | |
| "loss": 0.3133, | |
| "mean_token_accuracy": 0.8991359509527683, | |
| "num_tokens": 2617407.0, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.7710843373493976, | |
| "eval_loss": 0.8773419857025146, | |
| "eval_mean_token_accuracy": 0.7701640319824219, | |
| "eval_num_tokens": 2617407.0, | |
| "eval_runtime": 15.3576, | |
| "eval_samples_per_second": 25.785, | |
| "eval_steps_per_second": 3.256, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.8096385542168675, | |
| "grad_norm": 0.6024803881303014, | |
| "learning_rate": 9.036013927926048e-05, | |
| "loss": 0.3159, | |
| "mean_token_accuracy": 0.8986932337284088, | |
| "num_tokens": 2748479.0, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.8481927710843373, | |
| "grad_norm": 0.5729277681737998, | |
| "learning_rate": 8.943212079175391e-05, | |
| "loss": 0.3221, | |
| "mean_token_accuracy": 0.8965635932981968, | |
| "num_tokens": 2879551.0, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.8867469879518072, | |
| "grad_norm": 0.5396224216192379, | |
| "learning_rate": 8.846669854914396e-05, | |
| "loss": 0.3021, | |
| "mean_token_accuracy": 0.9026504307985306, | |
| "num_tokens": 3010185.0, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.9253012048192771, | |
| "grad_norm": 0.9017187841847952, | |
| "learning_rate": 8.746478831290648e-05, | |
| "loss": 0.3196, | |
| "mean_token_accuracy": 0.8984031714498997, | |
| "num_tokens": 3141257.0, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.963855421686747, | |
| "grad_norm": 0.5546422040544483, | |
| "learning_rate": 8.642734045558952e-05, | |
| "loss": 0.3107, | |
| "mean_token_accuracy": 0.8998587317764759, | |
| "num_tokens": 3271788.0, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.0192771084337349, | |
| "grad_norm": 0.7471418531311891, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 0.3844, | |
| "mean_token_accuracy": 0.9197667509317398, | |
| "num_tokens": 3435628.0, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 1.0578313253012048, | |
| "grad_norm": 0.5550574692250374, | |
| "learning_rate": 8.424980098237903e-05, | |
| "loss": 0.1996, | |
| "mean_token_accuracy": 0.9376297779381275, | |
| "num_tokens": 3566700.0, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 1.0963855421686748, | |
| "grad_norm": 0.5436657475279828, | |
| "learning_rate": 8.311177489457652e-05, | |
| "loss": 0.1868, | |
| "mean_token_accuracy": 0.9405608959496021, | |
| "num_tokens": 3697772.0, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 1.1349397590361445, | |
| "grad_norm": 0.6272110061503527, | |
| "learning_rate": 8.194234028259806e-05, | |
| "loss": 0.1832, | |
| "mean_token_accuracy": 0.9414692372083664, | |
| "num_tokens": 3828844.0, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 1.1734939759036145, | |
| "grad_norm": 0.7208474396560837, | |
| "learning_rate": 8.074260642600964e-05, | |
| "loss": 0.1907, | |
| "mean_token_accuracy": 0.9406677596271038, | |
| "num_tokens": 3959916.0, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.1734939759036145, | |
| "eval_loss": 1.0315335988998413, | |
| "eval_mean_token_accuracy": 0.7694289243221283, | |
| "eval_num_tokens": 3959916.0, | |
| "eval_runtime": 15.3983, | |
| "eval_samples_per_second": 25.717, | |
| "eval_steps_per_second": 3.247, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.2120481927710842, | |
| "grad_norm": 0.5340294943255592, | |
| "learning_rate": 7.951371134504599e-05, | |
| "loss": 0.1874, | |
| "mean_token_accuracy": 0.9399169348180294, | |
| "num_tokens": 4089879.0, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 1.2506024096385542, | |
| "grad_norm": 0.49194990095790314, | |
| "learning_rate": 7.82568207211296e-05, | |
| "loss": 0.1789, | |
| "mean_token_accuracy": 0.9420951530337334, | |
| "num_tokens": 4220951.0, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 1.2891566265060241, | |
| "grad_norm": 0.5057482409920033, | |
| "learning_rate": 7.697312679115125e-05, | |
| "loss": 0.1914, | |
| "mean_token_accuracy": 0.9384541548788548, | |
| "num_tokens": 4352023.0, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 1.3277108433734939, | |
| "grad_norm": 0.5052879188219443, | |
| "learning_rate": 7.566384721656104e-05, | |
| "loss": 0.1921, | |
| "mean_token_accuracy": 0.9380801320075989, | |
| "num_tokens": 4483095.0, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 1.3662650602409638, | |
| "grad_norm": 0.4730878422209722, | |
| "learning_rate": 7.433022392834282e-05, | |
| "loss": 0.1798, | |
| "mean_token_accuracy": 0.9419348575174809, | |
| "num_tokens": 4614167.0, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.4048192771084338, | |
| "grad_norm": 0.4938186363111339, | |
| "learning_rate": 7.297352194896739e-05, | |
| "loss": 0.1834, | |
| "mean_token_accuracy": 0.9405150972306728, | |
| "num_tokens": 4745239.0, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 1.4433734939759035, | |
| "grad_norm": 0.4723926832313344, | |
| "learning_rate": 7.159502819244206e-05, | |
| "loss": 0.1759, | |
| "mean_token_accuracy": 0.9420875199139118, | |
| "num_tokens": 4876311.0, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 1.4819277108433735, | |
| "grad_norm": 0.47675060498809807, | |
| "learning_rate": 7.019605024359474e-05, | |
| "loss": 0.1762, | |
| "mean_token_accuracy": 0.9431943222880363, | |
| "num_tokens": 5007383.0, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 1.5204819277108435, | |
| "grad_norm": 0.5364629489131902, | |
| "learning_rate": 6.877791511775063e-05, | |
| "loss": 0.1946, | |
| "mean_token_accuracy": 0.937858771532774, | |
| "num_tokens": 5138455.0, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 1.5590361445783132, | |
| "grad_norm": 0.46131419483606007, | |
| "learning_rate": 6.734196800197762e-05, | |
| "loss": 0.1776, | |
| "mean_token_accuracy": 0.9426218383014202, | |
| "num_tokens": 5269527.0, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.5590361445783132, | |
| "eval_loss": 0.9503161311149597, | |
| "eval_mean_token_accuracy": 0.7741586315631866, | |
| "eval_num_tokens": 5269527.0, | |
| "eval_runtime": 15.4787, | |
| "eval_samples_per_second": 25.583, | |
| "eval_steps_per_second": 3.23, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.5975903614457831, | |
| "grad_norm": 0.4621402991267078, | |
| "learning_rate": 6.588957097909508e-05, | |
| "loss": 0.171, | |
| "mean_token_accuracy": 0.9440052397549152, | |
| "num_tokens": 5400161.0, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 1.636144578313253, | |
| "grad_norm": 0.45927341036581576, | |
| "learning_rate": 6.442210173565561e-05, | |
| "loss": 0.1731, | |
| "mean_token_accuracy": 0.943843137472868, | |
| "num_tokens": 5531233.0, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 1.6746987951807228, | |
| "grad_norm": 0.47893034239528387, | |
| "learning_rate": 6.294095225512603e-05, | |
| "loss": 0.1851, | |
| "mean_token_accuracy": 0.939041905105114, | |
| "num_tokens": 5662305.0, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 1.7132530120481928, | |
| "grad_norm": 0.44140785586917847, | |
| "learning_rate": 6.14475274975067e-05, | |
| "loss": 0.1641, | |
| "mean_token_accuracy": 0.9456039071083069, | |
| "num_tokens": 5792219.0, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 1.7518072289156628, | |
| "grad_norm": 0.4592603098467548, | |
| "learning_rate": 5.9943244066641834e-05, | |
| "loss": 0.1833, | |
| "mean_token_accuracy": 0.9403319023549557, | |
| "num_tokens": 5923291.0, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.7903614457831325, | |
| "grad_norm": 0.47212047739990715, | |
| "learning_rate": 5.842952886648496e-05, | |
| "loss": 0.1793, | |
| "mean_token_accuracy": 0.94189178571105, | |
| "num_tokens": 6053822.0, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 1.8289156626506025, | |
| "grad_norm": 0.48517190308170305, | |
| "learning_rate": 5.6907817747594116e-05, | |
| "loss": 0.1859, | |
| "mean_token_accuracy": 0.9397441521286964, | |
| "num_tokens": 6184894.0, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 1.8674698795180724, | |
| "grad_norm": 0.46838562819306884, | |
| "learning_rate": 5.5379554145140574e-05, | |
| "loss": 0.1809, | |
| "mean_token_accuracy": 0.9421783201396465, | |
| "num_tokens": 6315149.0, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 1.9060240963855422, | |
| "grad_norm": 0.42945679090859756, | |
| "learning_rate": 5.38461877097232e-05, | |
| "loss": 0.1701, | |
| "mean_token_accuracy": 0.9449529275298119, | |
| "num_tokens": 6445345.0, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 1.944578313253012, | |
| "grad_norm": 0.42846414463088445, | |
| "learning_rate": 5.230917293228699e-05, | |
| "loss": 0.1708, | |
| "mean_token_accuracy": 0.9431876949965954, | |
| "num_tokens": 6576344.0, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.944578313253012, | |
| "eval_loss": 0.9409365653991699, | |
| "eval_mean_token_accuracy": 0.7767563450336457, | |
| "eval_num_tokens": 6576344.0, | |
| "eval_runtime": 15.3935, | |
| "eval_samples_per_second": 25.725, | |
| "eval_steps_per_second": 3.248, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.983132530120482, | |
| "grad_norm": 0.4374858491113943, | |
| "learning_rate": 5.0769967764450345e-05, | |
| "loss": 0.1789, | |
| "mean_token_accuracy": 0.9410112500190735, | |
| "num_tokens": 6707416.0, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 2.0385542168674697, | |
| "grad_norm": 0.3710697715007513, | |
| "learning_rate": 4.9230032235549667e-05, | |
| "loss": 0.1669, | |
| "mean_token_accuracy": 0.968594291806221, | |
| "num_tokens": 6871256.0, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 2.07710843373494, | |
| "grad_norm": 0.34301117426332123, | |
| "learning_rate": 4.7690827067713035e-05, | |
| "loss": 0.0901, | |
| "mean_token_accuracy": 0.9723079577088356, | |
| "num_tokens": 7002255.0, | |
| "step": 106 | |
| }, | |
| { | |
| "epoch": 2.1156626506024097, | |
| "grad_norm": 0.38494369189673533, | |
| "learning_rate": 4.6153812290276813e-05, | |
| "loss": 0.0896, | |
| "mean_token_accuracy": 0.9726428985595703, | |
| "num_tokens": 7133327.0, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 2.1542168674698794, | |
| "grad_norm": 0.4005604231123245, | |
| "learning_rate": 4.462044585485944e-05, | |
| "loss": 0.0795, | |
| "mean_token_accuracy": 0.9749339744448662, | |
| "num_tokens": 7263523.0, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 2.1927710843373496, | |
| "grad_norm": 0.5245059146895238, | |
| "learning_rate": 4.30921822524059e-05, | |
| "loss": 0.0832, | |
| "mean_token_accuracy": 0.9740702919661999, | |
| "num_tokens": 7394595.0, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 2.2313253012048193, | |
| "grad_norm": 0.44596089009324835, | |
| "learning_rate": 4.157047113351504e-05, | |
| "loss": 0.0837, | |
| "mean_token_accuracy": 0.9737268015742302, | |
| "num_tokens": 7525667.0, | |
| "step": 114 | |
| }, | |
| { | |
| "epoch": 2.269879518072289, | |
| "grad_norm": 0.4175768339270136, | |
| "learning_rate": 4.0056755933358184e-05, | |
| "loss": 0.0749, | |
| "mean_token_accuracy": 0.9757572114467621, | |
| "num_tokens": 7656739.0, | |
| "step": 116 | |
| }, | |
| { | |
| "epoch": 2.3084337349397592, | |
| "grad_norm": 1.443216067674743, | |
| "learning_rate": 3.855247250249331e-05, | |
| "loss": 0.0843, | |
| "mean_token_accuracy": 0.9733146131038666, | |
| "num_tokens": 7787811.0, | |
| "step": 118 | |
| }, | |
| { | |
| "epoch": 2.346987951807229, | |
| "grad_norm": 0.4402356237524311, | |
| "learning_rate": 3.705904774487396e-05, | |
| "loss": 0.0876, | |
| "mean_token_accuracy": 0.9715204574167728, | |
| "num_tokens": 7916616.0, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 2.346987951807229, | |
| "eval_loss": 1.1975072622299194, | |
| "eval_mean_token_accuracy": 0.7743996143341064, | |
| "eval_num_tokens": 7916616.0, | |
| "eval_runtime": 15.4013, | |
| "eval_samples_per_second": 25.712, | |
| "eval_steps_per_second": 3.246, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 2.3855421686746987, | |
| "grad_norm": 18.209390069415743, | |
| "learning_rate": 3.557789826434439e-05, | |
| "loss": 0.1444, | |
| "mean_token_accuracy": 0.9624297805130482, | |
| "num_tokens": 8047688.0, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 2.4240963855421684, | |
| "grad_norm": 70.14284982492848, | |
| "learning_rate": 3.411042902090492e-05, | |
| "loss": 0.1095, | |
| "mean_token_accuracy": 0.9678416661918163, | |
| "num_tokens": 8178760.0, | |
| "step": 124 | |
| }, | |
| { | |
| "epoch": 2.4626506024096386, | |
| "grad_norm": 0.6613555181637225, | |
| "learning_rate": 3.265803199802237e-05, | |
| "loss": 0.0922, | |
| "mean_token_accuracy": 0.9707498848438263, | |
| "num_tokens": 8309832.0, | |
| "step": 126 | |
| }, | |
| { | |
| "epoch": 2.5012048192771084, | |
| "grad_norm": 2.365906725464608, | |
| "learning_rate": 3.1222084882249375e-05, | |
| "loss": 0.0817, | |
| "mean_token_accuracy": 0.9744519479572773, | |
| "num_tokens": 8440904.0, | |
| "step": 128 | |
| }, | |
| { | |
| "epoch": 2.539759036144578, | |
| "grad_norm": 0.43221733799654627, | |
| "learning_rate": 2.980394975640526e-05, | |
| "loss": 0.0797, | |
| "mean_token_accuracy": 0.9748412370681763, | |
| "num_tokens": 8571976.0, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 2.5783132530120483, | |
| "grad_norm": 0.4001290128179629, | |
| "learning_rate": 2.8404971807557957e-05, | |
| "loss": 0.0806, | |
| "mean_token_accuracy": 0.9744519479572773, | |
| "num_tokens": 8703048.0, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 2.616867469879518, | |
| "grad_norm": 130.438666230395, | |
| "learning_rate": 2.7026478051032623e-05, | |
| "loss": 0.1193, | |
| "mean_token_accuracy": 0.9647807851433754, | |
| "num_tokens": 8834120.0, | |
| "step": 134 | |
| }, | |
| { | |
| "epoch": 2.6554216867469878, | |
| "grad_norm": 0.3714177013699221, | |
| "learning_rate": 2.5669776071657192e-05, | |
| "loss": 0.0772, | |
| "mean_token_accuracy": 0.9752766788005829, | |
| "num_tokens": 8964375.0, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 2.693975903614458, | |
| "grad_norm": 0.40141866230512896, | |
| "learning_rate": 2.4336152783438982e-05, | |
| "loss": 0.0804, | |
| "mean_token_accuracy": 0.9751916863024235, | |
| "num_tokens": 9094906.0, | |
| "step": 138 | |
| }, | |
| { | |
| "epoch": 2.7325301204819277, | |
| "grad_norm": 0.36374110805518306, | |
| "learning_rate": 2.302687320884876e-05, | |
| "loss": 0.0761, | |
| "mean_token_accuracy": 0.9760014712810516, | |
| "num_tokens": 9225978.0, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 2.7325301204819277, | |
| "eval_loss": 1.1796313524246216, | |
| "eval_mean_token_accuracy": 0.7758874070644378, | |
| "eval_num_tokens": 9225978.0, | |
| "eval_runtime": 15.4083, | |
| "eval_samples_per_second": 25.7, | |
| "eval_steps_per_second": 3.245, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 2.7710843373493974, | |
| "grad_norm": 0.40173825485325604, | |
| "learning_rate": 2.1743179278870407e-05, | |
| "loss": 0.0822, | |
| "mean_token_accuracy": 0.9739099964499474, | |
| "num_tokens": 9357050.0, | |
| "step": 142 | |
| }, | |
| { | |
| "epoch": 2.8096385542168676, | |
| "grad_norm": 0.4068099325110446, | |
| "learning_rate": 2.0486288654954028e-05, | |
| "loss": 0.0836, | |
| "mean_token_accuracy": 0.9737573340535164, | |
| "num_tokens": 9488122.0, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 2.8481927710843373, | |
| "grad_norm": 0.36122426413189357, | |
| "learning_rate": 1.925739357399038e-05, | |
| "loss": 0.0761, | |
| "mean_token_accuracy": 0.975909873843193, | |
| "num_tokens": 9619194.0, | |
| "step": 146 | |
| }, | |
| { | |
| "epoch": 2.886746987951807, | |
| "grad_norm": 0.3683534947273215, | |
| "learning_rate": 1.8057659717401947e-05, | |
| "loss": 0.0801, | |
| "mean_token_accuracy": 0.9750244319438934, | |
| "num_tokens": 9750266.0, | |
| "step": 148 | |
| }, | |
| { | |
| "epoch": 2.9253012048192772, | |
| "grad_norm": 0.30765914265523675, | |
| "learning_rate": 1.6888225105423507e-05, | |
| "loss": 0.072, | |
| "mean_token_accuracy": 0.9771159067749977, | |
| "num_tokens": 9881338.0, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 2.963855421686747, | |
| "grad_norm": 0.33714168632725033, | |
| "learning_rate": 1.575019901762097e-05, | |
| "loss": 0.0685, | |
| "mean_token_accuracy": 0.9783135503530502, | |
| "num_tokens": 10011972.0, | |
| "step": 152 | |
| }, | |
| { | |
| "epoch": 3.019277108433735, | |
| "grad_norm": 0.5470758567891826, | |
| "learning_rate": 1.4644660940672627e-05, | |
| "loss": 0.0965, | |
| "mean_token_accuracy": 0.9814179331064224, | |
| "num_tokens": 10175812.0, | |
| "step": 154 | |
| }, | |
| { | |
| "epoch": 3.057831325301205, | |
| "grad_norm": 0.2539257010993833, | |
| "learning_rate": 1.3572659544410494e-05, | |
| "loss": 0.0428, | |
| "mean_token_accuracy": 0.9880618005990982, | |
| "num_tokens": 10306884.0, | |
| "step": 156 | |
| }, | |
| { | |
| "epoch": 3.0963855421686746, | |
| "grad_norm": 0.25702092282460676, | |
| "learning_rate": 1.2535211687093535e-05, | |
| "loss": 0.0401, | |
| "mean_token_accuracy": 0.988520622253418, | |
| "num_tokens": 10437883.0, | |
| "step": 158 | |
| }, | |
| { | |
| "epoch": 3.1349397590361447, | |
| "grad_norm": 0.24086186857294117, | |
| "learning_rate": 1.1533301450856054e-05, | |
| "loss": 0.038, | |
| "mean_token_accuracy": 0.9894433952867985, | |
| "num_tokens": 10568955.0, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 3.1349397590361447, | |
| "eval_loss": 1.252247929573059, | |
| "eval_mean_token_accuracy": 0.7753590083122254, | |
| "eval_num_tokens": 10568955.0, | |
| "eval_runtime": 15.4135, | |
| "eval_samples_per_second": 25.692, | |
| "eval_steps_per_second": 3.244, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 3.1734939759036145, | |
| "grad_norm": 0.23198431830189042, | |
| "learning_rate": 1.0567879208246084e-05, | |
| "loss": 0.0379, | |
| "mean_token_accuracy": 0.9894204959273338, | |
| "num_tokens": 10700027.0, | |
| "step": 162 | |
| }, | |
| { | |
| "epoch": 3.212048192771084, | |
| "grad_norm": 0.24685981870387033, | |
| "learning_rate": 9.639860720739525e-06, | |
| "loss": 0.0368, | |
| "mean_token_accuracy": 0.9892983660101891, | |
| "num_tokens": 10831099.0, | |
| "step": 164 | |
| }, | |
| { | |
| "epoch": 3.2506024096385544, | |
| "grad_norm": 0.24529143344991325, | |
| "learning_rate": 8.75012627008489e-06, | |
| "loss": 0.0371, | |
| "mean_token_accuracy": 0.9891686029732227, | |
| "num_tokens": 10962171.0, | |
| "step": 166 | |
| }, | |
| { | |
| "epoch": 3.289156626506024, | |
| "grad_norm": 0.2613391302462127, | |
| "learning_rate": 7.899519823302743e-06, | |
| "loss": 0.0357, | |
| "mean_token_accuracy": 0.9897222742438316, | |
| "num_tokens": 11092367.0, | |
| "step": 168 | |
| }, | |
| { | |
| "epoch": 3.327710843373494, | |
| "grad_norm": 0.2562462479545936, | |
| "learning_rate": 7.088848232131861e-06, | |
| "loss": 0.0339, | |
| "mean_token_accuracy": 0.9898555837571621, | |
| "num_tokens": 11223439.0, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 3.3662650602409636, | |
| "grad_norm": 1.2788057307784748, | |
| "learning_rate": 6.318880467681526e-06, | |
| "loss": 0.0524, | |
| "mean_token_accuracy": 0.9860619232058525, | |
| "num_tokens": 11354511.0, | |
| "step": 172 | |
| }, | |
| { | |
| "epoch": 3.404819277108434, | |
| "grad_norm": 2.7968025902009024, | |
| "learning_rate": 5.590346891015758e-06, | |
| "loss": 0.0505, | |
| "mean_token_accuracy": 0.9857871308922768, | |
| "num_tokens": 11485583.0, | |
| "step": 174 | |
| }, | |
| { | |
| "epoch": 3.4433734939759035, | |
| "grad_norm": 0.2911939058631187, | |
| "learning_rate": 4.903938560361698e-06, | |
| "loss": 0.0369, | |
| "mean_token_accuracy": 0.9891422875225544, | |
| "num_tokens": 11616217.0, | |
| "step": 176 | |
| }, | |
| { | |
| "epoch": 3.4819277108433733, | |
| "grad_norm": 0.27968061605136607, | |
| "learning_rate": 4.260306575598949e-06, | |
| "loss": 0.0335, | |
| "mean_token_accuracy": 0.9900845773518085, | |
| "num_tokens": 11747289.0, | |
| "step": 178 | |
| }, | |
| { | |
| "epoch": 3.5204819277108435, | |
| "grad_norm": 0.2975123179246258, | |
| "learning_rate": 3.660061460651981e-06, | |
| "loss": 0.0366, | |
| "mean_token_accuracy": 0.9893365316092968, | |
| "num_tokens": 11878361.0, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 3.5204819277108435, | |
| "eval_loss": 1.428124189376831, | |
| "eval_mean_token_accuracy": 0.7740725183486938, | |
| "eval_num_tokens": 11878361.0, | |
| "eval_runtime": 15.3922, | |
| "eval_samples_per_second": 25.727, | |
| "eval_steps_per_second": 3.248, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 3.559036144578313, | |
| "grad_norm": 0.28680553389316427, | |
| "learning_rate": 3.1037725843711062e-06, | |
| "loss": 0.0412, | |
| "mean_token_accuracy": 0.9881381317973137, | |
| "num_tokens": 12009433.0, | |
| "step": 182 | |
| }, | |
| { | |
| "epoch": 3.597590361445783, | |
| "grad_norm": 0.2562230343512981, | |
| "learning_rate": 2.591967620451707e-06, | |
| "loss": 0.0339, | |
| "mean_token_accuracy": 0.9899166487157345, | |
| "num_tokens": 12140505.0, | |
| "step": 184 | |
| }, | |
| { | |
| "epoch": 3.636144578313253, | |
| "grad_norm": 0.28160755939439697, | |
| "learning_rate": 2.1251320469037827e-06, | |
| "loss": 0.0367, | |
| "mean_token_accuracy": 0.989214401692152, | |
| "num_tokens": 12271577.0, | |
| "step": 186 | |
| }, | |
| { | |
| "epoch": 3.674698795180723, | |
| "grad_norm": 0.26930385966588183, | |
| "learning_rate": 1.70370868554659e-06, | |
| "loss": 0.0395, | |
| "mean_token_accuracy": 0.9886953495442867, | |
| "num_tokens": 12402649.0, | |
| "step": 188 | |
| }, | |
| { | |
| "epoch": 3.7132530120481926, | |
| "grad_norm": 0.38341841164741197, | |
| "learning_rate": 1.328097281965357e-06, | |
| "loss": 0.0442, | |
| "mean_token_accuracy": 0.9869473651051521, | |
| "num_tokens": 12533721.0, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 3.7518072289156628, | |
| "grad_norm": 0.2788745994890486, | |
| "learning_rate": 9.986541263284077e-07, | |
| "loss": 0.0351, | |
| "mean_token_accuracy": 0.9894618764519691, | |
| "num_tokens": 12663435.0, | |
| "step": 192 | |
| }, | |
| { | |
| "epoch": 3.7903614457831325, | |
| "grad_norm": 0.5754887641552524, | |
| "learning_rate": 7.156917154243048e-07, | |
| "loss": 0.0424, | |
| "mean_token_accuracy": 0.9875961802899837, | |
| "num_tokens": 12794507.0, | |
| "step": 194 | |
| }, | |
| { | |
| "epoch": 3.8289156626506022, | |
| "grad_norm": 0.2887642296582216, | |
| "learning_rate": 4.794784562397458e-07, | |
| "loss": 0.0385, | |
| "mean_token_accuracy": 0.9885808527469635, | |
| "num_tokens": 12925579.0, | |
| "step": 196 | |
| }, | |
| { | |
| "epoch": 3.8674698795180724, | |
| "grad_norm": 0.2564012178920548, | |
| "learning_rate": 2.902384113592782e-07, | |
| "loss": 0.0347, | |
| "mean_token_accuracy": 0.9900998435914516, | |
| "num_tokens": 13056651.0, | |
| "step": 198 | |
| }, | |
| { | |
| "epoch": 3.906024096385542, | |
| "grad_norm": 0.2790701993318459, | |
| "learning_rate": 1.481510864283553e-07, | |
| "loss": 0.0364, | |
| "mean_token_accuracy": 0.9894052296876907, | |
| "num_tokens": 13187723.0, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 3.906024096385542, | |
| "eval_loss": 1.4324020147323608, | |
| "eval_mean_token_accuracy": 0.774030442237854, | |
| "eval_num_tokens": 13187723.0, | |
| "eval_runtime": 15.3847, | |
| "eval_samples_per_second": 25.74, | |
| "eval_steps_per_second": 3.25, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 3.944578313253012, | |
| "grad_norm": 0.26558018443477116, | |
| "learning_rate": 5.3351259881379014e-08, | |
| "loss": 0.0413, | |
| "mean_token_accuracy": 0.9883847609162331, | |
| "num_tokens": 13317686.0, | |
| "step": 202 | |
| }, | |
| { | |
| "epoch": 3.983132530120482, | |
| "grad_norm": 1.1597568643807872, | |
| "learning_rate": 5.928855096154484e-09, | |
| "loss": 0.0393, | |
| "mean_token_accuracy": 0.9885882064700127, | |
| "num_tokens": 13447600.0, | |
| "step": 204 | |
| }, | |
| { | |
| "epoch": 3.983132530120482, | |
| "step": 204, | |
| "total_flos": 19515995848704.0, | |
| "train_loss": 0.1621243642168302, | |
| "train_runtime": 1327.0646, | |
| "train_samples_per_second": 5.004, | |
| "train_steps_per_second": 0.154 | |
| } | |
| ], | |
| "logging_steps": 2, | |
| "max_steps": 204, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 4, | |
| "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": 19515995848704.0, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |