| { | |
| "best_global_step": null, | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 7.983132530120482, | |
| "eval_steps": 20, | |
| "global_step": 408, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.03855421686746988, | |
| "grad_norm": 1.3406128184465234, | |
| "learning_rate": 9.999851776425575e-06, | |
| "loss": 0.2215, | |
| "mean_token_accuracy": 0.9319889023900032, | |
| "num_tokens": 131072.0, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.07710843373493977, | |
| "grad_norm": 1.0868872715450462, | |
| "learning_rate": 9.998666040558187e-06, | |
| "loss": 0.2017, | |
| "mean_token_accuracy": 0.9355688355863094, | |
| "num_tokens": 262144.0, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.11566265060240964, | |
| "grad_norm": 0.8046336700273372, | |
| "learning_rate": 9.996294850025658e-06, | |
| "loss": 0.1993, | |
| "mean_token_accuracy": 0.9353551082313061, | |
| "num_tokens": 393216.0, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.15421686746987953, | |
| "grad_norm": 0.6584361140812891, | |
| "learning_rate": 9.992738767165791e-06, | |
| "loss": 0.1778, | |
| "mean_token_accuracy": 0.9412478767335415, | |
| "num_tokens": 524288.0, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.1927710843373494, | |
| "grad_norm": 0.6774335920541141, | |
| "learning_rate": 9.987998635318586e-06, | |
| "loss": 0.1902, | |
| "mean_token_accuracy": 0.9374810568988323, | |
| "num_tokens": 654484.0, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.23132530120481928, | |
| "grad_norm": 0.6105295086669309, | |
| "learning_rate": 9.982075578626235e-06, | |
| "loss": 0.1966, | |
| "mean_token_accuracy": 0.9350955821573734, | |
| "num_tokens": 785556.0, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.26987951807228916, | |
| "grad_norm": 0.6597942712481647, | |
| "learning_rate": 9.974971001766534e-06, | |
| "loss": 0.1967, | |
| "mean_token_accuracy": 0.9350586608052254, | |
| "num_tokens": 915519.0, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.30843373493975906, | |
| "grad_norm": 0.6435049991004593, | |
| "learning_rate": 9.96668658961975e-06, | |
| "loss": 0.1941, | |
| "mean_token_accuracy": 0.9362405501306057, | |
| "num_tokens": 1046591.0, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.3469879518072289, | |
| "grad_norm": 0.5815050330318928, | |
| "learning_rate": 9.957224306869053e-06, | |
| "loss": 0.1978, | |
| "mean_token_accuracy": 0.9344849325716496, | |
| "num_tokens": 1177663.0, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.3855421686746988, | |
| "grad_norm": 0.6062885715121378, | |
| "learning_rate": 9.946586397534572e-06, | |
| "loss": 0.1901, | |
| "mean_token_accuracy": 0.9375763460993767, | |
| "num_tokens": 1308735.0, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.3855421686746988, | |
| "eval_loss": 0.3250451982021332, | |
| "eval_mean_token_accuracy": 0.9015540636588479, | |
| "eval_num_tokens": 1308735.0, | |
| "eval_runtime": 32.894, | |
| "eval_samples_per_second": 25.993, | |
| "eval_steps_per_second": 3.253, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.42409638554216866, | |
| "grad_norm": 0.6159276714953087, | |
| "learning_rate": 9.93477538444123e-06, | |
| "loss": 0.1792, | |
| "mean_token_accuracy": 0.9396754540503025, | |
| "num_tokens": 1439807.0, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.46265060240963857, | |
| "grad_norm": 0.5991422890793899, | |
| "learning_rate": 9.92179406862043e-06, | |
| "loss": 0.1898, | |
| "mean_token_accuracy": 0.9368689768016338, | |
| "num_tokens": 1570062.0, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.5012048192771085, | |
| "grad_norm": 0.6010392974067564, | |
| "learning_rate": 9.907645528645791e-06, | |
| "loss": 0.1823, | |
| "mean_token_accuracy": 0.9397899508476257, | |
| "num_tokens": 1701134.0, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.5397590361445783, | |
| "grad_norm": 0.6463308628240576, | |
| "learning_rate": 9.892333119903045e-06, | |
| "loss": 0.1801, | |
| "mean_token_accuracy": 0.9400751106441021, | |
| "num_tokens": 1832133.0, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.5783132530120482, | |
| "grad_norm": 0.5855780926666729, | |
| "learning_rate": 9.875860473794302e-06, | |
| "loss": 0.1887, | |
| "mean_token_accuracy": 0.9366145730018616, | |
| "num_tokens": 1963205.0, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.6168674698795181, | |
| "grad_norm": 0.5767450209509862, | |
| "learning_rate": 9.85823149687683e-06, | |
| "loss": 0.1806, | |
| "mean_token_accuracy": 0.9403013698756695, | |
| "num_tokens": 2094277.0, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.655421686746988, | |
| "grad_norm": 0.6078338694602209, | |
| "learning_rate": 9.839450369936615e-06, | |
| "loss": 0.1804, | |
| "mean_token_accuracy": 0.9400189444422722, | |
| "num_tokens": 2225349.0, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.6939759036144578, | |
| "grad_norm": 0.5890734557721334, | |
| "learning_rate": 9.819521546996864e-06, | |
| "loss": 0.1824, | |
| "mean_token_accuracy": 0.9395788721740246, | |
| "num_tokens": 2355263.0, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.7325301204819277, | |
| "grad_norm": 31.323491593405347, | |
| "learning_rate": 9.798449754261716e-06, | |
| "loss": 0.2134, | |
| "mean_token_accuracy": 0.9329964704811573, | |
| "num_tokens": 2486335.0, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.7710843373493976, | |
| "grad_norm": 0.668421005752788, | |
| "learning_rate": 9.776239988995401e-06, | |
| "loss": 0.1756, | |
| "mean_token_accuracy": 0.9409272857010365, | |
| "num_tokens": 2617407.0, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.7710843373493976, | |
| "eval_loss": 0.323132187128067, | |
| "eval_mean_token_accuracy": 0.9019459669835099, | |
| "eval_num_tokens": 2617407.0, | |
| "eval_runtime": 32.8659, | |
| "eval_samples_per_second": 26.015, | |
| "eval_steps_per_second": 3.256, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.8096385542168675, | |
| "grad_norm": 0.6550725405812086, | |
| "learning_rate": 9.752897518337117e-06, | |
| "loss": 0.176, | |
| "mean_token_accuracy": 0.9411181136965752, | |
| "num_tokens": 2748479.0, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.8481927710843373, | |
| "grad_norm": 0.6735202725511205, | |
| "learning_rate": 9.72842787805191e-06, | |
| "loss": 0.1854, | |
| "mean_token_accuracy": 0.9372099563479424, | |
| "num_tokens": 2879551.0, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.8867469879518072, | |
| "grad_norm": 0.6034625832801067, | |
| "learning_rate": 9.702836871217838e-06, | |
| "loss": 0.1783, | |
| "mean_token_accuracy": 0.9399880617856979, | |
| "num_tokens": 3010185.0, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.9253012048192771, | |
| "grad_norm": 0.5972724356437912, | |
| "learning_rate": 9.676130566849757e-06, | |
| "loss": 0.1878, | |
| "mean_token_accuracy": 0.936996228992939, | |
| "num_tokens": 3141257.0, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.963855421686747, | |
| "grad_norm": 0.6007518988428301, | |
| "learning_rate": 9.64831529846001e-06, | |
| "loss": 0.1824, | |
| "mean_token_accuracy": 0.9391285218298435, | |
| "num_tokens": 3271788.0, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.0192771084337349, | |
| "grad_norm": 0.9787825024544956, | |
| "learning_rate": 9.619397662556434e-06, | |
| "loss": 0.2482, | |
| "mean_token_accuracy": 0.9454934179782868, | |
| "num_tokens": 3435628.0, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 1.0578313253012048, | |
| "grad_norm": 0.7523664235798198, | |
| "learning_rate": 9.589384517077945e-06, | |
| "loss": 0.1415, | |
| "mean_token_accuracy": 0.9537432938814163, | |
| "num_tokens": 3566700.0, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 1.0963855421686748, | |
| "grad_norm": 0.6132260326219375, | |
| "learning_rate": 9.558282979768164e-06, | |
| "loss": 0.1356, | |
| "mean_token_accuracy": 0.9553309828042984, | |
| "num_tokens": 3697772.0, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 1.1349397590361445, | |
| "grad_norm": 0.659449619068034, | |
| "learning_rate": 9.52610042648741e-06, | |
| "loss": 0.14, | |
| "mean_token_accuracy": 0.9534532353281975, | |
| "num_tokens": 3828844.0, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 1.1734939759036145, | |
| "grad_norm": 0.798456158560065, | |
| "learning_rate": 9.492844489463486e-06, | |
| "loss": 0.1402, | |
| "mean_token_accuracy": 0.9540104530751705, | |
| "num_tokens": 3959916.0, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.1734939759036145, | |
| "eval_loss": 0.3658570647239685, | |
| "eval_mean_token_accuracy": 0.9004586478260076, | |
| "eval_num_tokens": 3959916.0, | |
| "eval_runtime": 32.9568, | |
| "eval_samples_per_second": 25.943, | |
| "eval_steps_per_second": 3.247, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.2120481927710842, | |
| "grad_norm": 0.7106569795117357, | |
| "learning_rate": 9.458523055481658e-06, | |
| "loss": 0.1328, | |
| "mean_token_accuracy": 0.9561785310506821, | |
| "num_tokens": 4089879.0, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 1.2506024096385542, | |
| "grad_norm": 0.622759262640549, | |
| "learning_rate": 9.423144264014278e-06, | |
| "loss": 0.1269, | |
| "mean_token_accuracy": 0.9574377238750458, | |
| "num_tokens": 4220951.0, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 1.2891566265060241, | |
| "grad_norm": 0.6303844538287657, | |
| "learning_rate": 9.386716505290467e-06, | |
| "loss": 0.1413, | |
| "mean_token_accuracy": 0.9541173167526722, | |
| "num_tokens": 4352023.0, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 1.3277108433734939, | |
| "grad_norm": 0.6507157381170863, | |
| "learning_rate": 9.349248418306347e-06, | |
| "loss": 0.1455, | |
| "mean_token_accuracy": 0.9517892152070999, | |
| "num_tokens": 4483095.0, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 1.3662650602409638, | |
| "grad_norm": 0.602439194952957, | |
| "learning_rate": 9.310748888776254e-06, | |
| "loss": 0.1309, | |
| "mean_token_accuracy": 0.9569797366857529, | |
| "num_tokens": 4614167.0, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.4048192771084338, | |
| "grad_norm": 0.603487797008034, | |
| "learning_rate": 9.271227047025462e-06, | |
| "loss": 0.1333, | |
| "mean_token_accuracy": 0.9561706259846687, | |
| "num_tokens": 4745239.0, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 1.4433734939759035, | |
| "grad_norm": 0.6146915599817614, | |
| "learning_rate": 9.230692265824888e-06, | |
| "loss": 0.1333, | |
| "mean_token_accuracy": 0.9552088528871536, | |
| "num_tokens": 4876311.0, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 1.4819277108433735, | |
| "grad_norm": 0.6588127818916016, | |
| "learning_rate": 9.189154158168293e-06, | |
| "loss": 0.1308, | |
| "mean_token_accuracy": 0.9567278437316418, | |
| "num_tokens": 5007383.0, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 1.5204819277108435, | |
| "grad_norm": 0.6120032825642568, | |
| "learning_rate": 9.146622574992528e-06, | |
| "loss": 0.1461, | |
| "mean_token_accuracy": 0.9520487412810326, | |
| "num_tokens": 5138455.0, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 1.5590361445783132, | |
| "grad_norm": 0.6485104767700068, | |
| "learning_rate": 9.103107602841341e-06, | |
| "loss": 0.1384, | |
| "mean_token_accuracy": 0.9545066058635712, | |
| "num_tokens": 5269527.0, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.5590361445783132, | |
| "eval_loss": 0.3478308916091919, | |
| "eval_mean_token_accuracy": 0.9005198361717652, | |
| "eval_num_tokens": 5269527.0, | |
| "eval_runtime": 32.9211, | |
| "eval_samples_per_second": 25.971, | |
| "eval_steps_per_second": 3.25, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.5975903614457831, | |
| "grad_norm": 0.6036456378189476, | |
| "learning_rate": 9.058619561473308e-06, | |
| "loss": 0.1351, | |
| "mean_token_accuracy": 0.9550928063690662, | |
| "num_tokens": 5400161.0, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 1.636144578313253, | |
| "grad_norm": 0.5662228136563036, | |
| "learning_rate": 9.013169001414458e-06, | |
| "loss": 0.1389, | |
| "mean_token_accuracy": 0.9542012810707092, | |
| "num_tokens": 5531233.0, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 1.6746987951807228, | |
| "grad_norm": 0.668176616317804, | |
| "learning_rate": 8.966766701456177e-06, | |
| "loss": 0.1472, | |
| "mean_token_accuracy": 0.9509572051465511, | |
| "num_tokens": 5662305.0, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 1.7132530120481928, | |
| "grad_norm": 0.6659519140143432, | |
| "learning_rate": 8.91942366609897e-06, | |
| "loss": 0.1278, | |
| "mean_token_accuracy": 0.9569377563893795, | |
| "num_tokens": 5792219.0, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 1.7518072289156628, | |
| "grad_norm": 0.5730839087389755, | |
| "learning_rate": 8.871151122942692e-06, | |
| "loss": 0.1383, | |
| "mean_token_accuracy": 0.9540486186742783, | |
| "num_tokens": 5923291.0, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.7903614457831325, | |
| "grad_norm": 0.7339967333799423, | |
| "learning_rate": 8.821960520023884e-06, | |
| "loss": 0.1473, | |
| "mean_token_accuracy": 0.9519094601273537, | |
| "num_tokens": 6053822.0, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 1.8289156626506025, | |
| "grad_norm": 0.6406836390946679, | |
| "learning_rate": 8.771863523100821e-06, | |
| "loss": 0.1412, | |
| "mean_token_accuracy": 0.9523235335946083, | |
| "num_tokens": 6184894.0, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 1.8674698795180724, | |
| "grad_norm": 0.5600258442949189, | |
| "learning_rate": 8.720872012886918e-06, | |
| "loss": 0.1477, | |
| "mean_token_accuracy": 0.9504381828010082, | |
| "num_tokens": 6315149.0, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 1.9060240963855422, | |
| "grad_norm": 0.5933525309833521, | |
| "learning_rate": 8.668998082233186e-06, | |
| "loss": 0.1356, | |
| "mean_token_accuracy": 0.9551424346864223, | |
| "num_tokens": 6445345.0, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 1.944578313253012, | |
| "grad_norm": 0.6535174387269015, | |
| "learning_rate": 8.616254033260351e-06, | |
| "loss": 0.1365, | |
| "mean_token_accuracy": 0.9537569470703602, | |
| "num_tokens": 6576344.0, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.944578313253012, | |
| "eval_loss": 0.34725868701934814, | |
| "eval_mean_token_accuracy": 0.9006962408529264, | |
| "eval_num_tokens": 6576344.0, | |
| "eval_runtime": 32.8904, | |
| "eval_samples_per_second": 25.995, | |
| "eval_steps_per_second": 3.253, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.983132530120482, | |
| "grad_norm": 0.660872027341127, | |
| "learning_rate": 8.56265237444135e-06, | |
| "loss": 0.1433, | |
| "mean_token_accuracy": 0.9518960788846016, | |
| "num_tokens": 6707416.0, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 2.0385542168674697, | |
| "grad_norm": 0.6405367004816473, | |
| "learning_rate": 8.508205817634908e-06, | |
| "loss": 0.1641, | |
| "mean_token_accuracy": 0.9666707456111908, | |
| "num_tokens": 6871256.0, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 2.07710843373494, | |
| "grad_norm": 0.6396095656248479, | |
| "learning_rate": 8.452927275070858e-06, | |
| "loss": 0.1026, | |
| "mean_token_accuracy": 0.968053013086319, | |
| "num_tokens": 7002255.0, | |
| "step": 106 | |
| }, | |
| { | |
| "epoch": 2.1156626506024097, | |
| "grad_norm": 0.6728841385013591, | |
| "learning_rate": 8.39682985628795e-06, | |
| "loss": 0.1013, | |
| "mean_token_accuracy": 0.9683454521000385, | |
| "num_tokens": 7133327.0, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 2.1542168674698794, | |
| "grad_norm": 0.6356701289090003, | |
| "learning_rate": 8.339926865024871e-06, | |
| "loss": 0.1015, | |
| "mean_token_accuracy": 0.9669992625713348, | |
| "num_tokens": 7263523.0, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 2.1927710843373496, | |
| "grad_norm": 0.6909824113954701, | |
| "learning_rate": 8.282231796065215e-06, | |
| "loss": 0.0982, | |
| "mean_token_accuracy": 0.9683454521000385, | |
| "num_tokens": 7394595.0, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 2.2313253012048193, | |
| "grad_norm": 0.6949675615015238, | |
| "learning_rate": 8.223758332037121e-06, | |
| "loss": 0.0971, | |
| "mean_token_accuracy": 0.9685210138559341, | |
| "num_tokens": 7525667.0, | |
| "step": 114 | |
| }, | |
| { | |
| "epoch": 2.269879518072289, | |
| "grad_norm": 0.5752622256768609, | |
| "learning_rate": 8.164520340168404e-06, | |
| "loss": 0.1046, | |
| "mean_token_accuracy": 0.9659944511950016, | |
| "num_tokens": 7656739.0, | |
| "step": 116 | |
| }, | |
| { | |
| "epoch": 2.3084337349397592, | |
| "grad_norm": 0.6456834319746314, | |
| "learning_rate": 8.104531868997858e-06, | |
| "loss": 0.1, | |
| "mean_token_accuracy": 0.9683759845793247, | |
| "num_tokens": 7787811.0, | |
| "step": 118 | |
| }, | |
| { | |
| "epoch": 2.346987951807229, | |
| "grad_norm": 0.6268887677724969, | |
| "learning_rate": 8.043807145043604e-06, | |
| "loss": 0.1089, | |
| "mean_token_accuracy": 0.9643549546599388, | |
| "num_tokens": 7916616.0, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 2.346987951807229, | |
| "eval_loss": 0.3908001780509949, | |
| "eval_mean_token_accuracy": 0.8989515037180107, | |
| "eval_num_tokens": 7916616.0, | |
| "eval_runtime": 32.9535, | |
| "eval_samples_per_second": 25.946, | |
| "eval_steps_per_second": 3.247, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 2.3855421686746987, | |
| "grad_norm": 0.594448184955674, | |
| "learning_rate": 7.982360569429206e-06, | |
| "loss": 0.0919, | |
| "mean_token_accuracy": 0.9701926670968533, | |
| "num_tokens": 8047688.0, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 2.4240963855421684, | |
| "grad_norm": 0.6509482507816816, | |
| "learning_rate": 7.920206714468383e-06, | |
| "loss": 0.1, | |
| "mean_token_accuracy": 0.9679943285882473, | |
| "num_tokens": 8178760.0, | |
| "step": 124 | |
| }, | |
| { | |
| "epoch": 2.4626506024096386, | |
| "grad_norm": 0.6114017480150069, | |
| "learning_rate": 7.857360320209126e-06, | |
| "loss": 0.1079, | |
| "mean_token_accuracy": 0.9651777073740959, | |
| "num_tokens": 8309832.0, | |
| "step": 126 | |
| }, | |
| { | |
| "epoch": 2.5012048192771084, | |
| "grad_norm": 0.6679792242974475, | |
| "learning_rate": 7.793836290938026e-06, | |
| "loss": 0.0967, | |
| "mean_token_accuracy": 0.9691850952804089, | |
| "num_tokens": 8440904.0, | |
| "step": 128 | |
| }, | |
| { | |
| "epoch": 2.539759036144578, | |
| "grad_norm": 0.5886521368506863, | |
| "learning_rate": 7.729649691645673e-06, | |
| "loss": 0.0991, | |
| "mean_token_accuracy": 0.9681088253855705, | |
| "num_tokens": 8571976.0, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 2.5783132530120483, | |
| "grad_norm": 0.6059968735763466, | |
| "learning_rate": 7.664815744453918e-06, | |
| "loss": 0.1017, | |
| "mean_token_accuracy": 0.9668340943753719, | |
| "num_tokens": 8703048.0, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 2.616867469879518, | |
| "grad_norm": 0.63703209859714, | |
| "learning_rate": 7.599349825005892e-06, | |
| "loss": 0.1013, | |
| "mean_token_accuracy": 0.9671317860484123, | |
| "num_tokens": 8834120.0, | |
| "step": 134 | |
| }, | |
| { | |
| "epoch": 2.6554216867469878, | |
| "grad_norm": 0.6404376584659325, | |
| "learning_rate": 7.533267458819597e-06, | |
| "loss": 0.1081, | |
| "mean_token_accuracy": 0.9647228345274925, | |
| "num_tokens": 8964375.0, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 2.693975903614458, | |
| "grad_norm": 0.5731515889981167, | |
| "learning_rate": 7.466584317605978e-06, | |
| "loss": 0.0988, | |
| "mean_token_accuracy": 0.9682356528937817, | |
| "num_tokens": 9094906.0, | |
| "step": 138 | |
| }, | |
| { | |
| "epoch": 2.7325301204819277, | |
| "grad_norm": 0.5848185748214219, | |
| "learning_rate": 7.399316215552296e-06, | |
| "loss": 0.1016, | |
| "mean_token_accuracy": 0.9662845097482204, | |
| "num_tokens": 9225978.0, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 2.7325301204819277, | |
| "eval_loss": 0.3856795132160187, | |
| "eval_mean_token_accuracy": 0.8990027659407286, | |
| "eval_num_tokens": 9225978.0, | |
| "eval_runtime": 32.897, | |
| "eval_samples_per_second": 25.99, | |
| "eval_steps_per_second": 3.253, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 2.7710843373493974, | |
| "grad_norm": 0.6029958264244917, | |
| "learning_rate": 7.33147910557174e-06, | |
| "loss": 0.0958, | |
| "mean_token_accuracy": 0.9687576405704021, | |
| "num_tokens": 9357050.0, | |
| "step": 142 | |
| }, | |
| { | |
| "epoch": 2.8096385542168676, | |
| "grad_norm": 0.6783830228161134, | |
| "learning_rate": 7.26308907552012e-06, | |
| "loss": 0.1078, | |
| "mean_token_accuracy": 0.9650097787380219, | |
| "num_tokens": 9488122.0, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 2.8481927710843373, | |
| "grad_norm": 0.5827966276168525, | |
| "learning_rate": 7.194162344380561e-06, | |
| "loss": 0.1031, | |
| "mean_token_accuracy": 0.9667501300573349, | |
| "num_tokens": 9619194.0, | |
| "step": 146 | |
| }, | |
| { | |
| "epoch": 2.886746987951807, | |
| "grad_norm": 0.6450599697404067, | |
| "learning_rate": 7.124715258417111e-06, | |
| "loss": 0.1087, | |
| "mean_token_accuracy": 0.9652311392128468, | |
| "num_tokens": 9750266.0, | |
| "step": 148 | |
| }, | |
| { | |
| "epoch": 2.9253012048192772, | |
| "grad_norm": 0.5260422478675805, | |
| "learning_rate": 7.05476428729815e-06, | |
| "loss": 0.0976, | |
| "mean_token_accuracy": 0.9683836176991463, | |
| "num_tokens": 9881338.0, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 2.963855421686747, | |
| "grad_norm": 0.6011167506104717, | |
| "learning_rate": 6.984326020190544e-06, | |
| "loss": 0.0996, | |
| "mean_token_accuracy": 0.9671653471887112, | |
| "num_tokens": 10011972.0, | |
| "step": 152 | |
| }, | |
| { | |
| "epoch": 3.019277108433735, | |
| "grad_norm": 1.067138524628119, | |
| "learning_rate": 6.913417161825449e-06, | |
| "loss": 0.1396, | |
| "mean_token_accuracy": 0.9718551605939865, | |
| "num_tokens": 10175812.0, | |
| "step": 154 | |
| }, | |
| { | |
| "epoch": 3.057831325301205, | |
| "grad_norm": 0.6433380773043862, | |
| "learning_rate": 6.842054528536717e-06, | |
| "loss": 0.0748, | |
| "mean_token_accuracy": 0.9779708161950111, | |
| "num_tokens": 10306884.0, | |
| "step": 156 | |
| }, | |
| { | |
| "epoch": 3.0963855421686746, | |
| "grad_norm": 0.5518681598947663, | |
| "learning_rate": 6.770255044272826e-06, | |
| "loss": 0.0736, | |
| "mean_token_accuracy": 0.9776161313056946, | |
| "num_tokens": 10437883.0, | |
| "step": 158 | |
| }, | |
| { | |
| "epoch": 3.1349397590361447, | |
| "grad_norm": 0.5716170366461122, | |
| "learning_rate": 6.698035736583307e-06, | |
| "loss": 0.0704, | |
| "mean_token_accuracy": 0.9789860211312771, | |
| "num_tokens": 10568955.0, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 3.1349397590361447, | |
| "eval_loss": 0.4618055820465088, | |
| "eval_mean_token_accuracy": 0.8969022353118825, | |
| "eval_num_tokens": 10568955.0, | |
| "eval_runtime": 32.9015, | |
| "eval_samples_per_second": 25.987, | |
| "eval_steps_per_second": 3.252, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 3.1734939759036145, | |
| "grad_norm": 0.5660248629147534, | |
| "learning_rate": 6.625413732580577e-06, | |
| "loss": 0.0718, | |
| "mean_token_accuracy": 0.9778792187571526, | |
| "num_tokens": 10700027.0, | |
| "step": 162 | |
| }, | |
| { | |
| "epoch": 3.212048192771084, | |
| "grad_norm": 0.6184270631568148, | |
| "learning_rate": 6.552406254878175e-06, | |
| "loss": 0.0708, | |
| "mean_token_accuracy": 0.9779784493148327, | |
| "num_tokens": 10831099.0, | |
| "step": 164 | |
| }, | |
| { | |
| "epoch": 3.2506024096385544, | |
| "grad_norm": 0.7235816673040518, | |
| "learning_rate": 6.4790306175063535e-06, | |
| "loss": 0.0752, | |
| "mean_token_accuracy": 0.9767876826226711, | |
| "num_tokens": 10962171.0, | |
| "step": 166 | |
| }, | |
| { | |
| "epoch": 3.289156626506024, | |
| "grad_norm": 0.6381234994007693, | |
| "learning_rate": 6.405304221805972e-06, | |
| "loss": 0.0713, | |
| "mean_token_accuracy": 0.978339895606041, | |
| "num_tokens": 11092367.0, | |
| "step": 168 | |
| }, | |
| { | |
| "epoch": 3.327710843373494, | |
| "grad_norm": 0.625884194442065, | |
| "learning_rate": 6.331244552301705e-06, | |
| "loss": 0.0675, | |
| "mean_token_accuracy": 0.9787493944168091, | |
| "num_tokens": 11223439.0, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 3.3662650602409636, | |
| "grad_norm": 0.6347808690260591, | |
| "learning_rate": 6.2568691725555144e-06, | |
| "loss": 0.074, | |
| "mean_token_accuracy": 0.9773525334894657, | |
| "num_tokens": 11354511.0, | |
| "step": 172 | |
| }, | |
| { | |
| "epoch": 3.404819277108434, | |
| "grad_norm": 0.5756416236482357, | |
| "learning_rate": 6.182195721001366e-06, | |
| "loss": 0.0804, | |
| "mean_token_accuracy": 0.9753831885755062, | |
| "num_tokens": 11485583.0, | |
| "step": 174 | |
| }, | |
| { | |
| "epoch": 3.4433734939759035, | |
| "grad_norm": 0.5275947037323039, | |
| "learning_rate": 6.107241906762214e-06, | |
| "loss": 0.069, | |
| "mean_token_accuracy": 0.9786049015820026, | |
| "num_tokens": 11616217.0, | |
| "step": 176 | |
| }, | |
| { | |
| "epoch": 3.4819277108433733, | |
| "grad_norm": 0.5665166566187169, | |
| "learning_rate": 6.0320255054501985e-06, | |
| "loss": 0.0677, | |
| "mean_token_accuracy": 0.9791157841682434, | |
| "num_tokens": 11747289.0, | |
| "step": 178 | |
| }, | |
| { | |
| "epoch": 3.5204819277108435, | |
| "grad_norm": 0.6415976004443386, | |
| "learning_rate": 5.956564354951091e-06, | |
| "loss": 0.0704, | |
| "mean_token_accuracy": 0.9784059040248394, | |
| "num_tokens": 11878361.0, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 3.5204819277108435, | |
| "eval_loss": 0.45516592264175415, | |
| "eval_mean_token_accuracy": 0.8970169062926391, | |
| "eval_num_tokens": 11878361.0, | |
| "eval_runtime": 32.9299, | |
| "eval_samples_per_second": 25.964, | |
| "eval_steps_per_second": 3.249, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 3.559036144578313, | |
| "grad_norm": 0.5965252456073897, | |
| "learning_rate": 5.880876351193956e-06, | |
| "loss": 0.0719, | |
| "mean_token_accuracy": 0.9771998710930347, | |
| "num_tokens": 12009433.0, | |
| "step": 182 | |
| }, | |
| { | |
| "epoch": 3.597590361445783, | |
| "grad_norm": 0.5761543620705893, | |
| "learning_rate": 5.804979443907065e-06, | |
| "loss": 0.0768, | |
| "mean_token_accuracy": 0.9757114127278328, | |
| "num_tokens": 12140505.0, | |
| "step": 184 | |
| }, | |
| { | |
| "epoch": 3.636144578313253, | |
| "grad_norm": 0.6381827427058839, | |
| "learning_rate": 5.728891632361043e-06, | |
| "loss": 0.072, | |
| "mean_token_accuracy": 0.9778639525175095, | |
| "num_tokens": 12271577.0, | |
| "step": 186 | |
| }, | |
| { | |
| "epoch": 3.674698795180723, | |
| "grad_norm": 0.5760484373146149, | |
| "learning_rate": 5.65263096110026e-06, | |
| "loss": 0.0737, | |
| "mean_token_accuracy": 0.9771769717335701, | |
| "num_tokens": 12402649.0, | |
| "step": 188 | |
| }, | |
| { | |
| "epoch": 3.7132530120481926, | |
| "grad_norm": 0.6007872958623798, | |
| "learning_rate": 5.576215515663489e-06, | |
| "loss": 0.0771, | |
| "mean_token_accuracy": 0.9759022407233715, | |
| "num_tokens": 12533721.0, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 3.7518072289156628, | |
| "grad_norm": 0.5865088698200008, | |
| "learning_rate": 5.499663418294858e-06, | |
| "loss": 0.0765, | |
| "mean_token_accuracy": 0.9760418757796288, | |
| "num_tokens": 12663435.0, | |
| "step": 192 | |
| }, | |
| { | |
| "epoch": 3.7903614457831325, | |
| "grad_norm": 0.5869508573232244, | |
| "learning_rate": 5.4229928236460705e-06, | |
| "loss": 0.0688, | |
| "mean_token_accuracy": 0.9785509333014488, | |
| "num_tokens": 12794507.0, | |
| "step": 194 | |
| }, | |
| { | |
| "epoch": 3.8289156626506022, | |
| "grad_norm": 0.6151614870939663, | |
| "learning_rate": 5.346221914470959e-06, | |
| "loss": 0.076, | |
| "mean_token_accuracy": 0.9760396368801594, | |
| "num_tokens": 12925579.0, | |
| "step": 196 | |
| }, | |
| { | |
| "epoch": 3.8674698795180724, | |
| "grad_norm": 0.5387118456522078, | |
| "learning_rate": 5.2693688973133675e-06, | |
| "loss": 0.0705, | |
| "mean_token_accuracy": 0.9784669689834118, | |
| "num_tokens": 13056651.0, | |
| "step": 198 | |
| }, | |
| { | |
| "epoch": 3.906024096385542, | |
| "grad_norm": 0.5994849742352796, | |
| "learning_rate": 5.192451998189392e-06, | |
| "loss": 0.0756, | |
| "mean_token_accuracy": 0.9767113514244556, | |
| "num_tokens": 13187723.0, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 3.906024096385542, | |
| "eval_loss": 0.4474319517612457, | |
| "eval_mean_token_accuracy": 0.8968760477048214, | |
| "eval_num_tokens": 13187723.0, | |
| "eval_runtime": 32.9071, | |
| "eval_samples_per_second": 25.982, | |
| "eval_steps_per_second": 3.252, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 3.944578313253012, | |
| "grad_norm": 0.5535712673806997, | |
| "learning_rate": 5.115489458265006e-06, | |
| "loss": 0.0742, | |
| "mean_token_accuracy": 0.976105697453022, | |
| "num_tokens": 13317686.0, | |
| "step": 202 | |
| }, | |
| { | |
| "epoch": 3.983132530120482, | |
| "grad_norm": 0.5578205642046994, | |
| "learning_rate": 5.038499529530094e-06, | |
| "loss": 0.0706, | |
| "mean_token_accuracy": 0.9777835607528687, | |
| "num_tokens": 13447600.0, | |
| "step": 204 | |
| }, | |
| { | |
| "epoch": 4.03855421686747, | |
| "grad_norm": 0.5165401701064283, | |
| "learning_rate": 4.961500470469908e-06, | |
| "loss": 0.0864, | |
| "mean_token_accuracy": 0.9837689340114594, | |
| "num_tokens": 13611440.0, | |
| "step": 206 | |
| }, | |
| { | |
| "epoch": 4.0771084337349395, | |
| "grad_norm": 0.5606600779130405, | |
| "learning_rate": 4.8845105417349955e-06, | |
| "loss": 0.0514, | |
| "mean_token_accuracy": 0.9854289144277573, | |
| "num_tokens": 13740537.0, | |
| "step": 208 | |
| }, | |
| { | |
| "epoch": 4.11566265060241, | |
| "grad_norm": 0.5032380365598611, | |
| "learning_rate": 4.807548001810611e-06, | |
| "loss": 0.054, | |
| "mean_token_accuracy": 0.984481867402792, | |
| "num_tokens": 13871609.0, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 4.15421686746988, | |
| "grad_norm": 0.5627408303007229, | |
| "learning_rate": 4.730631102686635e-06, | |
| "loss": 0.0551, | |
| "mean_token_accuracy": 0.9837490878999233, | |
| "num_tokens": 14002681.0, | |
| "step": 212 | |
| }, | |
| { | |
| "epoch": 4.192771084337349, | |
| "grad_norm": 0.6584478813758422, | |
| "learning_rate": 4.653778085529043e-06, | |
| "loss": 0.0573, | |
| "mean_token_accuracy": 0.9833140000700951, | |
| "num_tokens": 14133753.0, | |
| "step": 214 | |
| }, | |
| { | |
| "epoch": 4.231325301204819, | |
| "grad_norm": 0.6355929893733159, | |
| "learning_rate": 4.577007176353931e-06, | |
| "loss": 0.053, | |
| "mean_token_accuracy": 0.9844788759946823, | |
| "num_tokens": 14264314.0, | |
| "step": 216 | |
| }, | |
| { | |
| "epoch": 4.2698795180722895, | |
| "grad_norm": 0.5944238331457758, | |
| "learning_rate": 4.5003365817051434e-06, | |
| "loss": 0.0495, | |
| "mean_token_accuracy": 0.9854665398597717, | |
| "num_tokens": 14395386.0, | |
| "step": 218 | |
| }, | |
| { | |
| "epoch": 4.308433734939759, | |
| "grad_norm": 0.5486764850427266, | |
| "learning_rate": 4.4237844843365126e-06, | |
| "loss": 0.0548, | |
| "mean_token_accuracy": 0.9835506267845631, | |
| "num_tokens": 14526458.0, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 4.308433734939759, | |
| "eval_loss": 0.5238527059555054, | |
| "eval_mean_token_accuracy": 0.895290755222891, | |
| "eval_num_tokens": 14526458.0, | |
| "eval_runtime": 32.8995, | |
| "eval_samples_per_second": 25.988, | |
| "eval_steps_per_second": 3.252, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 4.346987951807229, | |
| "grad_norm": 0.5193566382330146, | |
| "learning_rate": 4.347369038899744e-06, | |
| "loss": 0.0512, | |
| "mean_token_accuracy": 0.9847192876040936, | |
| "num_tokens": 14656421.0, | |
| "step": 222 | |
| }, | |
| { | |
| "epoch": 4.385542168674699, | |
| "grad_norm": 0.6081963974814661, | |
| "learning_rate": 4.271108367638959e-06, | |
| "loss": 0.0512, | |
| "mean_token_accuracy": 0.9849474877119064, | |
| "num_tokens": 14787493.0, | |
| "step": 224 | |
| }, | |
| { | |
| "epoch": 4.424096385542168, | |
| "grad_norm": 0.5352653032319554, | |
| "learning_rate": 4.195020556092935e-06, | |
| "loss": 0.0508, | |
| "mean_token_accuracy": 0.984955120831728, | |
| "num_tokens": 14918565.0, | |
| "step": 226 | |
| }, | |
| { | |
| "epoch": 4.462650602409639, | |
| "grad_norm": 0.5471233273653687, | |
| "learning_rate": 4.119123648806046e-06, | |
| "loss": 0.0524, | |
| "mean_token_accuracy": 0.9842299744486809, | |
| "num_tokens": 15049637.0, | |
| "step": 228 | |
| }, | |
| { | |
| "epoch": 4.501204819277109, | |
| "grad_norm": 0.5761419852713972, | |
| "learning_rate": 4.043435645048911e-06, | |
| "loss": 0.0538, | |
| "mean_token_accuracy": 0.9835735261440277, | |
| "num_tokens": 15180709.0, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 4.539759036144578, | |
| "grad_norm": 0.5193378395323519, | |
| "learning_rate": 3.967974494549803e-06, | |
| "loss": 0.0527, | |
| "mean_token_accuracy": 0.9835735261440277, | |
| "num_tokens": 15311781.0, | |
| "step": 232 | |
| }, | |
| { | |
| "epoch": 4.578313253012048, | |
| "grad_norm": 0.6838502479167475, | |
| "learning_rate": 3.892758093237788e-06, | |
| "loss": 0.0505, | |
| "mean_token_accuracy": 0.9845734648406506, | |
| "num_tokens": 15442853.0, | |
| "step": 234 | |
| }, | |
| { | |
| "epoch": 4.6168674698795185, | |
| "grad_norm": 0.6161795707876571, | |
| "learning_rate": 3.8178042789986355e-06, | |
| "loss": 0.0597, | |
| "mean_token_accuracy": 0.9818713404238224, | |
| "num_tokens": 15573925.0, | |
| "step": 236 | |
| }, | |
| { | |
| "epoch": 4.655421686746988, | |
| "grad_norm": 0.5673557334484985, | |
| "learning_rate": 3.743130827444487e-06, | |
| "loss": 0.0551, | |
| "mean_token_accuracy": 0.9827491492033005, | |
| "num_tokens": 15704997.0, | |
| "step": 238 | |
| }, | |
| { | |
| "epoch": 4.693975903614458, | |
| "grad_norm": 0.605102586233592, | |
| "learning_rate": 3.6687554476982954e-06, | |
| "loss": 0.0516, | |
| "mean_token_accuracy": 0.9845963642001152, | |
| "num_tokens": 15836069.0, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 4.693975903614458, | |
| "eval_loss": 0.5330836176872253, | |
| "eval_mean_token_accuracy": 0.894900638366414, | |
| "eval_num_tokens": 15836069.0, | |
| "eval_runtime": 32.8994, | |
| "eval_samples_per_second": 25.988, | |
| "eval_steps_per_second": 3.252, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 4.732530120481927, | |
| "grad_norm": 0.6102630121983813, | |
| "learning_rate": 3.5946957781940296e-06, | |
| "loss": 0.0592, | |
| "mean_token_accuracy": 0.9820545352995396, | |
| "num_tokens": 15967141.0, | |
| "step": 242 | |
| }, | |
| { | |
| "epoch": 4.771084337349397, | |
| "grad_norm": 0.5601462915097579, | |
| "learning_rate": 3.5209693824936486e-06, | |
| "loss": 0.049, | |
| "mean_token_accuracy": 0.9852528125047684, | |
| "num_tokens": 16098213.0, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 4.809638554216868, | |
| "grad_norm": 0.5729256978696017, | |
| "learning_rate": 3.4475937451218257e-06, | |
| "loss": 0.0534, | |
| "mean_token_accuracy": 0.9839246496558189, | |
| "num_tokens": 16229285.0, | |
| "step": 246 | |
| }, | |
| { | |
| "epoch": 4.848192771084337, | |
| "grad_norm": 0.5576976700355626, | |
| "learning_rate": 3.3745862674194246e-06, | |
| "loss": 0.053, | |
| "mean_token_accuracy": 0.984092578291893, | |
| "num_tokens": 16360357.0, | |
| "step": 248 | |
| }, | |
| { | |
| "epoch": 4.886746987951807, | |
| "grad_norm": 0.568301505042255, | |
| "learning_rate": 3.301964263416693e-06, | |
| "loss": 0.0501, | |
| "mean_token_accuracy": 0.984886422753334, | |
| "num_tokens": 16491429.0, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 4.925301204819277, | |
| "grad_norm": 0.506093754486047, | |
| "learning_rate": 3.2297449557271743e-06, | |
| "loss": 0.0503, | |
| "mean_token_accuracy": 0.9847882427275181, | |
| "num_tokens": 16621960.0, | |
| "step": 252 | |
| }, | |
| { | |
| "epoch": 4.9638554216867465, | |
| "grad_norm": 0.5314105489580034, | |
| "learning_rate": 3.1579454714632853e-06, | |
| "loss": 0.0527, | |
| "mean_token_accuracy": 0.9841104596853256, | |
| "num_tokens": 16752156.0, | |
| "step": 254 | |
| }, | |
| { | |
| "epoch": 5.019277108433735, | |
| "grad_norm": 0.8525557002128702, | |
| "learning_rate": 3.0865828381745515e-06, | |
| "loss": 0.0798, | |
| "mean_token_accuracy": 0.9847118079662323, | |
| "num_tokens": 16915558.0, | |
| "step": 256 | |
| }, | |
| { | |
| "epoch": 5.057831325301205, | |
| "grad_norm": 0.4807676519731547, | |
| "learning_rate": 3.015673979809457e-06, | |
| "loss": 0.0433, | |
| "mean_token_accuracy": 0.9880142249166965, | |
| "num_tokens": 17045521.0, | |
| "step": 258 | |
| }, | |
| { | |
| "epoch": 5.096385542168675, | |
| "grad_norm": 0.46177338047601046, | |
| "learning_rate": 2.9452357127018516e-06, | |
| "loss": 0.0409, | |
| "mean_token_accuracy": 0.9885750971734524, | |
| "num_tokens": 17175776.0, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 5.096385542168675, | |
| "eval_loss": 0.5568886995315552, | |
| "eval_mean_token_accuracy": 0.8947282572773015, | |
| "eval_num_tokens": 17175776.0, | |
| "eval_runtime": 32.8508, | |
| "eval_samples_per_second": 26.027, | |
| "eval_steps_per_second": 3.257, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 5.134939759036144, | |
| "grad_norm": 0.4354223067866842, | |
| "learning_rate": 2.8752847415828923e-06, | |
| "loss": 0.0378, | |
| "mean_token_accuracy": 0.9894357621669769, | |
| "num_tokens": 17306848.0, | |
| "step": 262 | |
| }, | |
| { | |
| "epoch": 5.1734939759036145, | |
| "grad_norm": 0.46752372012241933, | |
| "learning_rate": 2.80583765561944e-06, | |
| "loss": 0.0369, | |
| "mean_token_accuracy": 0.989603690803051, | |
| "num_tokens": 17437920.0, | |
| "step": 264 | |
| }, | |
| { | |
| "epoch": 5.212048192771085, | |
| "grad_norm": 0.495717384078659, | |
| "learning_rate": 2.736910924479881e-06, | |
| "loss": 0.0403, | |
| "mean_token_accuracy": 0.9886197298765182, | |
| "num_tokens": 17568116.0, | |
| "step": 266 | |
| }, | |
| { | |
| "epoch": 5.250602409638554, | |
| "grad_norm": 0.5445603960561214, | |
| "learning_rate": 2.668520894428259e-06, | |
| "loss": 0.0396, | |
| "mean_token_accuracy": 0.9885961189866066, | |
| "num_tokens": 17699188.0, | |
| "step": 268 | |
| }, | |
| { | |
| "epoch": 5.289156626506024, | |
| "grad_norm": 0.648469580339254, | |
| "learning_rate": 2.600683784447704e-06, | |
| "loss": 0.0398, | |
| "mean_token_accuracy": 0.9886266514658928, | |
| "num_tokens": 17830260.0, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 5.327710843373494, | |
| "grad_norm": 0.5496241291341013, | |
| "learning_rate": 2.5334156823940237e-06, | |
| "loss": 0.0434, | |
| "mean_token_accuracy": 0.9878862388432026, | |
| "num_tokens": 17961332.0, | |
| "step": 272 | |
| }, | |
| { | |
| "epoch": 5.366265060240964, | |
| "grad_norm": 0.5345216492493586, | |
| "learning_rate": 2.466732541180404e-06, | |
| "loss": 0.0425, | |
| "mean_token_accuracy": 0.9877641089260578, | |
| "num_tokens": 18092404.0, | |
| "step": 274 | |
| }, | |
| { | |
| "epoch": 5.404819277108434, | |
| "grad_norm": 0.48538502077655216, | |
| "learning_rate": 2.4006501749941097e-06, | |
| "loss": 0.04, | |
| "mean_token_accuracy": 0.9883747585117817, | |
| "num_tokens": 18223476.0, | |
| "step": 276 | |
| }, | |
| { | |
| "epoch": 5.443373493975904, | |
| "grad_norm": 0.4611939418614471, | |
| "learning_rate": 2.335184255546083e-06, | |
| "loss": 0.0406, | |
| "mean_token_accuracy": 0.9881762973964214, | |
| "num_tokens": 18354548.0, | |
| "step": 278 | |
| }, | |
| { | |
| "epoch": 5.481927710843373, | |
| "grad_norm": 0.5074811080316414, | |
| "learning_rate": 2.2703503083543288e-06, | |
| "loss": 0.0408, | |
| "mean_token_accuracy": 0.9885350540280342, | |
| "num_tokens": 18485620.0, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 5.481927710843373, | |
| "eval_loss": 0.5997635126113892, | |
| "eval_mean_token_accuracy": 0.893693176942451, | |
| "eval_num_tokens": 18485620.0, | |
| "eval_runtime": 32.9111, | |
| "eval_samples_per_second": 25.979, | |
| "eval_steps_per_second": 3.251, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 5.5204819277108435, | |
| "grad_norm": 0.4599599600334865, | |
| "learning_rate": 2.206163709061976e-06, | |
| "loss": 0.0411, | |
| "mean_token_accuracy": 0.988504521548748, | |
| "num_tokens": 18616692.0, | |
| "step": 282 | |
| }, | |
| { | |
| "epoch": 5.559036144578314, | |
| "grad_norm": 0.45053903346967306, | |
| "learning_rate": 2.1426396797908764e-06, | |
| "loss": 0.0373, | |
| "mean_token_accuracy": 0.9894662946462631, | |
| "num_tokens": 18747764.0, | |
| "step": 284 | |
| }, | |
| { | |
| "epoch": 5.597590361445783, | |
| "grad_norm": 0.5371349909088498, | |
| "learning_rate": 2.0797932855316183e-06, | |
| "loss": 0.0405, | |
| "mean_token_accuracy": 0.988229539245367, | |
| "num_tokens": 18877678.0, | |
| "step": 286 | |
| }, | |
| { | |
| "epoch": 5.636144578313253, | |
| "grad_norm": 0.4593266181875638, | |
| "learning_rate": 2.017639430570794e-06, | |
| "loss": 0.0412, | |
| "mean_token_accuracy": 0.9881075993180275, | |
| "num_tokens": 19008750.0, | |
| "step": 288 | |
| }, | |
| { | |
| "epoch": 5.674698795180722, | |
| "grad_norm": 0.48430078531022114, | |
| "learning_rate": 1.956192854956397e-06, | |
| "loss": 0.0415, | |
| "mean_token_accuracy": 0.9877486675977707, | |
| "num_tokens": 19139281.0, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 5.713253012048193, | |
| "grad_norm": 0.4989863180632785, | |
| "learning_rate": 1.8954681310021434e-06, | |
| "loss": 0.0424, | |
| "mean_token_accuracy": 0.9875351153314114, | |
| "num_tokens": 19270353.0, | |
| "step": 292 | |
| }, | |
| { | |
| "epoch": 5.751807228915663, | |
| "grad_norm": 0.4623064292722575, | |
| "learning_rate": 1.8354796598315977e-06, | |
| "loss": 0.0407, | |
| "mean_token_accuracy": 0.9884434565901756, | |
| "num_tokens": 19401425.0, | |
| "step": 294 | |
| }, | |
| { | |
| "epoch": 5.790361445783132, | |
| "grad_norm": 0.5014798249213933, | |
| "learning_rate": 1.7762416679628792e-06, | |
| "loss": 0.0394, | |
| "mean_token_accuracy": 0.9886740408837795, | |
| "num_tokens": 19532424.0, | |
| "step": 296 | |
| }, | |
| { | |
| "epoch": 5.828915662650602, | |
| "grad_norm": 0.5392913823526073, | |
| "learning_rate": 1.7177682039347875e-06, | |
| "loss": 0.0408, | |
| "mean_token_accuracy": 0.9878557063639164, | |
| "num_tokens": 19663496.0, | |
| "step": 298 | |
| }, | |
| { | |
| "epoch": 5.867469879518072, | |
| "grad_norm": 0.5613539090124229, | |
| "learning_rate": 1.6600731349751303e-06, | |
| "loss": 0.0396, | |
| "mean_token_accuracy": 0.988573219627142, | |
| "num_tokens": 19794568.0, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 5.867469879518072, | |
| "eval_loss": 0.6202346682548523, | |
| "eval_mean_token_accuracy": 0.8934637369396531, | |
| "eval_num_tokens": 19794568.0, | |
| "eval_runtime": 32.9094, | |
| "eval_samples_per_second": 25.98, | |
| "eval_steps_per_second": 3.251, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 5.906024096385542, | |
| "grad_norm": 0.6075220334585247, | |
| "learning_rate": 1.6031701437120512e-06, | |
| "loss": 0.0436, | |
| "mean_token_accuracy": 0.9871763586997986, | |
| "num_tokens": 19925640.0, | |
| "step": 302 | |
| }, | |
| { | |
| "epoch": 5.944578313253012, | |
| "grad_norm": 0.478268700720953, | |
| "learning_rate": 1.5470727249291423e-06, | |
| "loss": 0.0392, | |
| "mean_token_accuracy": 0.9886266514658928, | |
| "num_tokens": 20056712.0, | |
| "step": 304 | |
| }, | |
| { | |
| "epoch": 5.983132530120482, | |
| "grad_norm": 0.5299056008151652, | |
| "learning_rate": 1.4917941823650917e-06, | |
| "loss": 0.0393, | |
| "mean_token_accuracy": 0.9882984273135662, | |
| "num_tokens": 20187784.0, | |
| "step": 306 | |
| }, | |
| { | |
| "epoch": 6.03855421686747, | |
| "grad_norm": 0.4515619693414329, | |
| "learning_rate": 1.4373476255586515e-06, | |
| "loss": 0.0539, | |
| "mean_token_accuracy": 0.9903883755207061, | |
| "num_tokens": 20351624.0, | |
| "step": 308 | |
| }, | |
| { | |
| "epoch": 6.0771084337349395, | |
| "grad_norm": 0.4461331299908182, | |
| "learning_rate": 1.383745966739652e-06, | |
| "loss": 0.0377, | |
| "mean_token_accuracy": 0.9897639863193035, | |
| "num_tokens": 20482696.0, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 6.11566265060241, | |
| "grad_norm": 0.3811282983266442, | |
| "learning_rate": 1.3310019177668154e-06, | |
| "loss": 0.0335, | |
| "mean_token_accuracy": 0.9908020906150341, | |
| "num_tokens": 20613768.0, | |
| "step": 312 | |
| }, | |
| { | |
| "epoch": 6.15421686746988, | |
| "grad_norm": 0.4003286418611843, | |
| "learning_rate": 1.2791279871130824e-06, | |
| "loss": 0.0315, | |
| "mean_token_accuracy": 0.9916570000350475, | |
| "num_tokens": 20744840.0, | |
| "step": 314 | |
| }, | |
| { | |
| "epoch": 6.192771084337349, | |
| "grad_norm": 0.4136786703234865, | |
| "learning_rate": 1.2281364768991804e-06, | |
| "loss": 0.0357, | |
| "mean_token_accuracy": 0.990313570946455, | |
| "num_tokens": 20875912.0, | |
| "step": 316 | |
| }, | |
| { | |
| "epoch": 6.231325301204819, | |
| "grad_norm": 0.45632226149343463, | |
| "learning_rate": 1.1780394799761163e-06, | |
| "loss": 0.0324, | |
| "mean_token_accuracy": 0.9911643601953983, | |
| "num_tokens": 21006167.0, | |
| "step": 318 | |
| }, | |
| { | |
| "epoch": 6.2698795180722895, | |
| "grad_norm": 0.3996603424603635, | |
| "learning_rate": 1.1288488770573097e-06, | |
| "loss": 0.0305, | |
| "mean_token_accuracy": 0.9917750433087349, | |
| "num_tokens": 21135643.0, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 6.2698795180722895, | |
| "eval_loss": 0.6617009043693542, | |
| "eval_mean_token_accuracy": 0.8925447820503021, | |
| "eval_num_tokens": 21135643.0, | |
| "eval_runtime": 32.9359, | |
| "eval_samples_per_second": 25.96, | |
| "eval_steps_per_second": 3.249, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 6.308433734939759, | |
| "grad_norm": 0.4550106121984443, | |
| "learning_rate": 1.0805763339010329e-06, | |
| "loss": 0.0337, | |
| "mean_token_accuracy": 0.990831833332777, | |
| "num_tokens": 21266174.0, | |
| "step": 322 | |
| }, | |
| { | |
| "epoch": 6.346987951807229, | |
| "grad_norm": 0.4540741073756364, | |
| "learning_rate": 1.0332332985438248e-06, | |
| "loss": 0.0348, | |
| "mean_token_accuracy": 0.990550197660923, | |
| "num_tokens": 21397246.0, | |
| "step": 324 | |
| }, | |
| { | |
| "epoch": 6.385542168674699, | |
| "grad_norm": 0.47292748183552674, | |
| "learning_rate": 9.868309985855446e-07, | |
| "loss": 0.0326, | |
| "mean_token_accuracy": 0.9910158179700375, | |
| "num_tokens": 21528318.0, | |
| "step": 326 | |
| }, | |
| { | |
| "epoch": 6.424096385542168, | |
| "grad_norm": 0.46337369369151543, | |
| "learning_rate": 9.41380438526694e-07, | |
| "loss": 0.0328, | |
| "mean_token_accuracy": 0.990702860057354, | |
| "num_tokens": 21659390.0, | |
| "step": 328 | |
| }, | |
| { | |
| "epoch": 6.462650602409639, | |
| "grad_norm": 0.5096633620765385, | |
| "learning_rate": 8.968923971586596e-07, | |
| "loss": 0.0321, | |
| "mean_token_accuracy": 0.9912142790853977, | |
| "num_tokens": 21790462.0, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 6.501204819277109, | |
| "grad_norm": 0.4849844285466508, | |
| "learning_rate": 8.533774250074727e-07, | |
| "loss": 0.0307, | |
| "mean_token_accuracy": 0.991565402597189, | |
| "num_tokens": 21921534.0, | |
| "step": 332 | |
| }, | |
| { | |
| "epoch": 6.539759036144578, | |
| "grad_norm": 0.47614922479928773, | |
| "learning_rate": 8.108458418317089e-07, | |
| "loss": 0.0329, | |
| "mean_token_accuracy": 0.9907412827014923, | |
| "num_tokens": 22051497.0, | |
| "step": 334 | |
| }, | |
| { | |
| "epoch": 6.578313253012048, | |
| "grad_norm": 0.49852850537940474, | |
| "learning_rate": 7.693077341751138e-07, | |
| "loss": 0.0323, | |
| "mean_token_accuracy": 0.9908555224537849, | |
| "num_tokens": 22182569.0, | |
| "step": 336 | |
| }, | |
| { | |
| "epoch": 6.6168674698795185, | |
| "grad_norm": 0.4273359220675533, | |
| "learning_rate": 7.287729529745386e-07, | |
| "loss": 0.0328, | |
| "mean_token_accuracy": 0.9911041483283043, | |
| "num_tokens": 22312765.0, | |
| "step": 338 | |
| }, | |
| { | |
| "epoch": 6.655421686746988, | |
| "grad_norm": 0.4557886589116148, | |
| "learning_rate": 6.892511112237472e-07, | |
| "loss": 0.0344, | |
| "mean_token_accuracy": 0.9900464117527008, | |
| "num_tokens": 22443837.0, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 6.655421686746988, | |
| "eval_loss": 0.6737558841705322, | |
| "eval_mean_token_accuracy": 0.8922743284813711, | |
| "eval_num_tokens": 22443837.0, | |
| "eval_runtime": 32.8919, | |
| "eval_samples_per_second": 25.994, | |
| "eval_steps_per_second": 3.253, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 6.693975903614458, | |
| "grad_norm": 0.46620750191290955, | |
| "learning_rate": 6.507515816936538e-07, | |
| "loss": 0.0333, | |
| "mean_token_accuracy": 0.9908784218132496, | |
| "num_tokens": 22574909.0, | |
| "step": 342 | |
| }, | |
| { | |
| "epoch": 6.732530120481927, | |
| "grad_norm": 0.4697001737130722, | |
| "learning_rate": 6.132834947095334e-07, | |
| "loss": 0.0328, | |
| "mean_token_accuracy": 0.9907486587762833, | |
| "num_tokens": 22705981.0, | |
| "step": 344 | |
| }, | |
| { | |
| "epoch": 6.771084337349397, | |
| "grad_norm": 0.4430515622762748, | |
| "learning_rate": 5.768557359857241e-07, | |
| "loss": 0.0321, | |
| "mean_token_accuracy": 0.9912448115646839, | |
| "num_tokens": 22837053.0, | |
| "step": 346 | |
| }, | |
| { | |
| "epoch": 6.809638554216868, | |
| "grad_norm": 0.4361690495303729, | |
| "learning_rate": 5.414769445183432e-07, | |
| "loss": 0.0328, | |
| "mean_token_accuracy": 0.9909089542925358, | |
| "num_tokens": 22968125.0, | |
| "step": 348 | |
| }, | |
| { | |
| "epoch": 6.848192771084337, | |
| "grad_norm": 0.4200405293131056, | |
| "learning_rate": 5.071555105365156e-07, | |
| "loss": 0.0346, | |
| "mean_token_accuracy": 0.9901227429509163, | |
| "num_tokens": 23099197.0, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 6.886746987951807, | |
| "grad_norm": 0.47982475678883507, | |
| "learning_rate": 4.738995735125895e-07, | |
| "loss": 0.0358, | |
| "mean_token_accuracy": 0.9901380091905594, | |
| "num_tokens": 23230269.0, | |
| "step": 352 | |
| }, | |
| { | |
| "epoch": 6.925301204819277, | |
| "grad_norm": 0.4588352996826278, | |
| "learning_rate": 4.4171702023183663e-07, | |
| "loss": 0.0345, | |
| "mean_token_accuracy": 0.9904306195676327, | |
| "num_tokens": 23361268.0, | |
| "step": 354 | |
| }, | |
| { | |
| "epoch": 6.9638554216867465, | |
| "grad_norm": 0.49076163739088335, | |
| "learning_rate": 4.10615482922056e-07, | |
| "loss": 0.034, | |
| "mean_token_accuracy": 0.9904967658221722, | |
| "num_tokens": 23492340.0, | |
| "step": 356 | |
| }, | |
| { | |
| "epoch": 7.019277108433735, | |
| "grad_norm": 0.7037303292281402, | |
| "learning_rate": 3.8060233744356634e-07, | |
| "loss": 0.0476, | |
| "mean_token_accuracy": 0.9915913552045822, | |
| "num_tokens": 23656180.0, | |
| "step": 358 | |
| }, | |
| { | |
| "epoch": 7.057831325301205, | |
| "grad_norm": 0.46496724208097717, | |
| "learning_rate": 3.5168470153998937e-07, | |
| "loss": 0.0307, | |
| "mean_token_accuracy": 0.9916331619024277, | |
| "num_tokens": 23786143.0, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 7.057831325301205, | |
| "eval_loss": 0.6740805506706238, | |
| "eval_mean_token_accuracy": 0.8924460611610769, | |
| "eval_num_tokens": 23786143.0, | |
| "eval_runtime": 32.9163, | |
| "eval_samples_per_second": 25.975, | |
| "eval_steps_per_second": 3.251, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 7.096385542168675, | |
| "grad_norm": 0.43993509563002636, | |
| "learning_rate": 3.238694331502451e-07, | |
| "loss": 0.0299, | |
| "mean_token_accuracy": 0.9917791299521923, | |
| "num_tokens": 23917215.0, | |
| "step": 362 | |
| }, | |
| { | |
| "epoch": 7.134939759036144, | |
| "grad_norm": 0.42297111455766123, | |
| "learning_rate": 2.9716312878216194e-07, | |
| "loss": 0.0295, | |
| "mean_token_accuracy": 0.992290548980236, | |
| "num_tokens": 24048287.0, | |
| "step": 364 | |
| }, | |
| { | |
| "epoch": 7.1734939759036145, | |
| "grad_norm": 0.3740711012992464, | |
| "learning_rate": 2.71572121948091e-07, | |
| "loss": 0.0287, | |
| "mean_token_accuracy": 0.9923897795379162, | |
| "num_tokens": 24179359.0, | |
| "step": 366 | |
| }, | |
| { | |
| "epoch": 7.212048192771085, | |
| "grad_norm": 0.40083480465861504, | |
| "learning_rate": 2.471024816628836e-07, | |
| "loss": 0.0309, | |
| "mean_token_accuracy": 0.9913974739611149, | |
| "num_tokens": 24310431.0, | |
| "step": 368 | |
| }, | |
| { | |
| "epoch": 7.250602409638554, | |
| "grad_norm": 0.433443148283099, | |
| "learning_rate": 2.237600110046001e-07, | |
| "loss": 0.0325, | |
| "mean_token_accuracy": 0.9914738051593304, | |
| "num_tokens": 24441503.0, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 7.289156626506024, | |
| "grad_norm": 0.3913556728274319, | |
| "learning_rate": 2.0155024573828452e-07, | |
| "loss": 0.0309, | |
| "mean_token_accuracy": 0.9921867847442627, | |
| "num_tokens": 24572502.0, | |
| "step": 372 | |
| }, | |
| { | |
| "epoch": 7.327710843373494, | |
| "grad_norm": 0.3931756889237697, | |
| "learning_rate": 1.8047845300313726e-07, | |
| "loss": 0.032, | |
| "mean_token_accuracy": 0.991260077804327, | |
| "num_tokens": 24703574.0, | |
| "step": 374 | |
| }, | |
| { | |
| "epoch": 7.366265060240964, | |
| "grad_norm": 0.39372361510112325, | |
| "learning_rate": 1.6054963006338742e-07, | |
| "loss": 0.0307, | |
| "mean_token_accuracy": 0.9918172955513, | |
| "num_tokens": 24834646.0, | |
| "step": 376 | |
| }, | |
| { | |
| "epoch": 7.404819277108434, | |
| "grad_norm": 0.4539746897925293, | |
| "learning_rate": 1.4176850312317246e-07, | |
| "loss": 0.0286, | |
| "mean_token_accuracy": 0.9920920878648758, | |
| "num_tokens": 24965718.0, | |
| "step": 378 | |
| }, | |
| { | |
| "epoch": 7.443373493975904, | |
| "grad_norm": 0.3804093569086182, | |
| "learning_rate": 1.241395262056999e-07, | |
| "loss": 0.0282, | |
| "mean_token_accuracy": 0.9927180036902428, | |
| "num_tokens": 25096790.0, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 7.443373493975904, | |
| "eval_loss": 0.6863826513290405, | |
| "eval_mean_token_accuracy": 0.8920925671809188, | |
| "eval_num_tokens": 25096790.0, | |
| "eval_runtime": 32.9256, | |
| "eval_samples_per_second": 25.968, | |
| "eval_steps_per_second": 3.25, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 7.481927710843373, | |
| "grad_norm": 0.4086380049589114, | |
| "learning_rate": 1.0766688009695548e-07, | |
| "loss": 0.0296, | |
| "mean_token_accuracy": 0.9921302534639835, | |
| "num_tokens": 25227862.0, | |
| "step": 382 | |
| }, | |
| { | |
| "epoch": 7.5204819277108435, | |
| "grad_norm": 0.42260905745056304, | |
| "learning_rate": 9.235447135421127e-08, | |
| "loss": 0.03, | |
| "mean_token_accuracy": 0.9916875325143337, | |
| "num_tokens": 25358934.0, | |
| "step": 384 | |
| }, | |
| { | |
| "epoch": 7.559036144578314, | |
| "grad_norm": 0.4176362136377749, | |
| "learning_rate": 7.820593137957244e-08, | |
| "loss": 0.031, | |
| "mean_token_accuracy": 0.9918630942702293, | |
| "num_tokens": 25490006.0, | |
| "step": 386 | |
| }, | |
| { | |
| "epoch": 7.597590361445783, | |
| "grad_norm": 0.45701242241975604, | |
| "learning_rate": 6.522461555877213e-08, | |
| "loss": 0.0349, | |
| "mean_token_accuracy": 0.9904204346239567, | |
| "num_tokens": 25621078.0, | |
| "step": 388 | |
| }, | |
| { | |
| "epoch": 7.636144578313253, | |
| "grad_norm": 0.40879094458461246, | |
| "learning_rate": 5.341360246542804e-08, | |
| "loss": 0.0306, | |
| "mean_token_accuracy": 0.9920004904270172, | |
| "num_tokens": 25752150.0, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 7.674698795180722, | |
| "grad_norm": 0.43647524957553, | |
| "learning_rate": 4.2775693130948094e-08, | |
| "loss": 0.0317, | |
| "mean_token_accuracy": 0.9910463504493237, | |
| "num_tokens": 25883222.0, | |
| "step": 392 | |
| }, | |
| { | |
| "epoch": 7.713253012048193, | |
| "grad_norm": 0.41444591834134303, | |
| "learning_rate": 3.3313410380250157e-08, | |
| "loss": 0.0312, | |
| "mean_token_accuracy": 0.9916607923805714, | |
| "num_tokens": 26012936.0, | |
| "step": 394 | |
| }, | |
| { | |
| "epoch": 7.751807228915663, | |
| "grad_norm": 0.3803028538212038, | |
| "learning_rate": 2.5028998233467272e-08, | |
| "loss": 0.0306, | |
| "mean_token_accuracy": 0.9918249286711216, | |
| "num_tokens": 26144008.0, | |
| "step": 396 | |
| }, | |
| { | |
| "epoch": 7.790361445783132, | |
| "grad_norm": 0.3600455476234704, | |
| "learning_rate": 1.7924421373766153e-08, | |
| "loss": 0.0292, | |
| "mean_token_accuracy": 0.9922676496207714, | |
| "num_tokens": 26275080.0, | |
| "step": 398 | |
| }, | |
| { | |
| "epoch": 7.828915662650602, | |
| "grad_norm": 0.39321030231512055, | |
| "learning_rate": 1.200136468141544e-08, | |
| "loss": 0.0324, | |
| "mean_token_accuracy": 0.9913364090025425, | |
| "num_tokens": 26406152.0, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 7.828915662650602, | |
| "eval_loss": 0.6899031400680542, | |
| "eval_mean_token_accuracy": 0.892040293350398, | |
| "eval_num_tokens": 26406152.0, | |
| "eval_runtime": 32.9667, | |
| "eval_samples_per_second": 25.935, | |
| "eval_steps_per_second": 3.246, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 7.867469879518072, | |
| "grad_norm": 0.41654832211115, | |
| "learning_rate": 7.261232834209208e-09, | |
| "loss": 0.0292, | |
| "mean_token_accuracy": 0.9922164119780064, | |
| "num_tokens": 26536786.0, | |
| "step": 402 | |
| }, | |
| { | |
| "epoch": 7.906024096385542, | |
| "grad_norm": 0.40099638349259836, | |
| "learning_rate": 3.705149974342348e-09, | |
| "loss": 0.0307, | |
| "mean_token_accuracy": 0.9913211427628994, | |
| "num_tokens": 26667858.0, | |
| "step": 404 | |
| }, | |
| { | |
| "epoch": 7.944578313253012, | |
| "grad_norm": 0.42947674316382145, | |
| "learning_rate": 1.3339594418138036e-09, | |
| "loss": 0.0307, | |
| "mean_token_accuracy": 0.9916405789554119, | |
| "num_tokens": 26796896.0, | |
| "step": 406 | |
| }, | |
| { | |
| "epoch": 7.983132530120482, | |
| "grad_norm": 0.41338624680752933, | |
| "learning_rate": 1.4822357442656475e-10, | |
| "loss": 0.0293, | |
| "mean_token_accuracy": 0.992038656026125, | |
| "num_tokens": 26927968.0, | |
| "step": 408 | |
| }, | |
| { | |
| "epoch": 7.983132530120482, | |
| "step": 408, | |
| "total_flos": 39091383042048.0, | |
| "train_loss": 0.08385189656423879, | |
| "train_runtime": 3861.3463, | |
| "train_samples_per_second": 3.439, | |
| "train_steps_per_second": 0.106 | |
| } | |
| ], | |
| "logging_steps": 2, | |
| "max_steps": 408, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 8, | |
| "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": 39091383042048.0, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |