| { |
| "best_global_step": null, |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 0.2066115702479339, |
| "eval_steps": 500, |
| "global_step": 275, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.003756574004507889, |
| "grad_norm": 31.56857681274414, |
| "learning_rate": 7.904761904761904e-06, |
| "loss": 3.2699, |
| "mean_token_accuracy": 0.5073748350143432, |
| "num_tokens": 477077.0, |
| "step": 5 |
| }, |
| { |
| "epoch": 0.007513148009015778, |
| "grad_norm": 3.5605764389038086, |
| "learning_rate": 1.7785714285714282e-05, |
| "loss": 0.9697, |
| "mean_token_accuracy": 0.709954810142517, |
| "num_tokens": 965438.0, |
| "step": 10 |
| }, |
| { |
| "epoch": 0.011269722013523666, |
| "grad_norm": 1.1990835666656494, |
| "learning_rate": 2.7666666666666667e-05, |
| "loss": 0.2976, |
| "mean_token_accuracy": 0.8834105372428894, |
| "num_tokens": 1425790.0, |
| "step": 15 |
| }, |
| { |
| "epoch": 0.015026296018031555, |
| "grad_norm": 0.5824474692344666, |
| "learning_rate": 3.7547619047619045e-05, |
| "loss": 0.2422, |
| "mean_token_accuracy": 0.8969236493110657, |
| "num_tokens": 1891264.0, |
| "step": 20 |
| }, |
| { |
| "epoch": 0.018782870022539443, |
| "grad_norm": 0.5905170440673828, |
| "learning_rate": 4.7428571428571427e-05, |
| "loss": 0.2387, |
| "mean_token_accuracy": 0.8978369951248169, |
| "num_tokens": 2375491.0, |
| "step": 25 |
| }, |
| { |
| "epoch": 0.022539444027047332, |
| "grad_norm": 0.6168460249900818, |
| "learning_rate": 5.730952380952381e-05, |
| "loss": 0.2306, |
| "mean_token_accuracy": 0.9046724081039429, |
| "num_tokens": 2857161.0, |
| "step": 30 |
| }, |
| { |
| "epoch": 0.02629601803155522, |
| "grad_norm": 0.7918921113014221, |
| "learning_rate": 6.719047619047618e-05, |
| "loss": 0.2401, |
| "mean_token_accuracy": 0.8969136714935303, |
| "num_tokens": 3306156.0, |
| "step": 35 |
| }, |
| { |
| "epoch": 0.03005259203606311, |
| "grad_norm": 0.8585085868835449, |
| "learning_rate": 6.916544738388534e-05, |
| "loss": 0.2134, |
| "mean_token_accuracy": 0.9039913058280945, |
| "num_tokens": 3763533.0, |
| "step": 40 |
| }, |
| { |
| "epoch": 0.033809166040571, |
| "grad_norm": 0.28870654106140137, |
| "learning_rate": 6.916049424405181e-05, |
| "loss": 0.2211, |
| "mean_token_accuracy": 0.9063485980033874, |
| "num_tokens": 4244159.0, |
| "step": 45 |
| }, |
| { |
| "epoch": 0.037565740045078885, |
| "grad_norm": 0.29164770245552063, |
| "learning_rate": 6.915173176905315e-05, |
| "loss": 0.206, |
| "mean_token_accuracy": 0.9049365878105163, |
| "num_tokens": 4690542.0, |
| "step": 50 |
| }, |
| { |
| "epoch": 0.04132231404958678, |
| "grad_norm": 0.4590369760990143, |
| "learning_rate": 6.913916124610476e-05, |
| "loss": 0.2053, |
| "mean_token_accuracy": 0.9130877494812012, |
| "num_tokens": 5165891.0, |
| "step": 55 |
| }, |
| { |
| "epoch": 0.045078888054094664, |
| "grad_norm": 0.4134966731071472, |
| "learning_rate": 6.912278452182769e-05, |
| "loss": 0.194, |
| "mean_token_accuracy": 0.914238840341568, |
| "num_tokens": 5641202.0, |
| "step": 60 |
| }, |
| { |
| "epoch": 0.048835462058602556, |
| "grad_norm": 0.6215408444404602, |
| "learning_rate": 6.910260400197734e-05, |
| "loss": 0.2015, |
| "mean_token_accuracy": 0.9092130601406098, |
| "num_tokens": 6099554.0, |
| "step": 65 |
| }, |
| { |
| "epoch": 0.05259203606311044, |
| "grad_norm": 0.303625226020813, |
| "learning_rate": 6.907862265109009e-05, |
| "loss": 0.1796, |
| "mean_token_accuracy": 0.9203076422214508, |
| "num_tokens": 6567957.0, |
| "step": 70 |
| }, |
| { |
| "epoch": 0.056348610067618335, |
| "grad_norm": 0.460435152053833, |
| "learning_rate": 6.90508439920478e-05, |
| "loss": 0.1987, |
| "mean_token_accuracy": 0.9155427157878876, |
| "num_tokens": 7018607.0, |
| "step": 75 |
| }, |
| { |
| "epoch": 0.06010518407212622, |
| "grad_norm": 0.361350417137146, |
| "learning_rate": 6.901927210556026e-05, |
| "loss": 0.18, |
| "mean_token_accuracy": 0.9237741231918335, |
| "num_tokens": 7506189.0, |
| "step": 80 |
| }, |
| { |
| "epoch": 0.0638617580766341, |
| "grad_norm": 0.4573143422603607, |
| "learning_rate": 6.89839116295658e-05, |
| "loss": 0.1803, |
| "mean_token_accuracy": 0.9219691574573516, |
| "num_tokens": 7979553.0, |
| "step": 85 |
| }, |
| { |
| "epoch": 0.067618332081142, |
| "grad_norm": 0.4155818521976471, |
| "learning_rate": 6.894476775854987e-05, |
| "loss": 0.1824, |
| "mean_token_accuracy": 0.9162850677967072, |
| "num_tokens": 8432544.0, |
| "step": 90 |
| }, |
| { |
| "epoch": 0.07137490608564989, |
| "grad_norm": 0.36648187041282654, |
| "learning_rate": 6.890184624278208e-05, |
| "loss": 0.1697, |
| "mean_token_accuracy": 0.9244144022464752, |
| "num_tokens": 8908324.0, |
| "step": 95 |
| }, |
| { |
| "epoch": 0.07513148009015777, |
| "grad_norm": 0.2350839227437973, |
| "learning_rate": 6.88551533874714e-05, |
| "loss": 0.162, |
| "mean_token_accuracy": 0.9230327785015107, |
| "num_tokens": 9380778.0, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.07888805409466566, |
| "grad_norm": 0.2973095774650574, |
| "learning_rate": 6.880469605183995e-05, |
| "loss": 0.1715, |
| "mean_token_accuracy": 0.9241137385368348, |
| "num_tokens": 9845282.0, |
| "step": 105 |
| }, |
| { |
| "epoch": 0.08264462809917356, |
| "grad_norm": 0.5604135394096375, |
| "learning_rate": 6.875048164811534e-05, |
| "loss": 0.1613, |
| "mean_token_accuracy": 0.9276620626449585, |
| "num_tokens": 10323072.0, |
| "step": 110 |
| }, |
| { |
| "epoch": 0.08640120210368145, |
| "grad_norm": 0.3381125032901764, |
| "learning_rate": 6.869251814044191e-05, |
| "loss": 0.1655, |
| "mean_token_accuracy": 0.9268105328083038, |
| "num_tokens": 10797974.0, |
| "step": 115 |
| }, |
| { |
| "epoch": 0.09015777610818933, |
| "grad_norm": 0.23712225258350372, |
| "learning_rate": 6.863081404371065e-05, |
| "loss": 0.1629, |
| "mean_token_accuracy": 0.9290018200874328, |
| "num_tokens": 11246005.0, |
| "step": 120 |
| }, |
| { |
| "epoch": 0.09391435011269722, |
| "grad_norm": 0.25958552956581116, |
| "learning_rate": 6.856537842230844e-05, |
| "loss": 0.1618, |
| "mean_token_accuracy": 0.9229864597320556, |
| "num_tokens": 11713157.0, |
| "step": 125 |
| }, |
| { |
| "epoch": 0.09767092411720511, |
| "grad_norm": 0.4623725712299347, |
| "learning_rate": 6.849622088878647e-05, |
| "loss": 0.1625, |
| "mean_token_accuracy": 0.925396466255188, |
| "num_tokens": 12179816.0, |
| "step": 130 |
| }, |
| { |
| "epoch": 0.10142749812171299, |
| "grad_norm": 0.246930330991745, |
| "learning_rate": 6.842335160244813e-05, |
| "loss": 0.1497, |
| "mean_token_accuracy": 0.9329809546470642, |
| "num_tokens": 12644159.0, |
| "step": 135 |
| }, |
| { |
| "epoch": 0.10518407212622088, |
| "grad_norm": 0.34383425116539, |
| "learning_rate": 6.834678126785663e-05, |
| "loss": 0.1629, |
| "mean_token_accuracy": 0.9280821084976196, |
| "num_tokens": 13102450.0, |
| "step": 140 |
| }, |
| { |
| "epoch": 0.10894064613072878, |
| "grad_norm": 0.2808777689933777, |
| "learning_rate": 6.826652113326245e-05, |
| "loss": 0.149, |
| "mean_token_accuracy": 0.9282186388969421, |
| "num_tokens": 13578416.0, |
| "step": 145 |
| }, |
| { |
| "epoch": 0.11269722013523667, |
| "grad_norm": 0.2689949870109558, |
| "learning_rate": 6.8182582988951e-05, |
| "loss": 0.1543, |
| "mean_token_accuracy": 0.9298548936843872, |
| "num_tokens": 14043790.0, |
| "step": 150 |
| }, |
| { |
| "epoch": 0.11645379413974455, |
| "grad_norm": 0.45267754793167114, |
| "learning_rate": 6.809497916551061e-05, |
| "loss": 0.1584, |
| "mean_token_accuracy": 0.9262810707092285, |
| "num_tokens": 14509006.0, |
| "step": 155 |
| }, |
| { |
| "epoch": 0.12021036814425244, |
| "grad_norm": 0.23879091441631317, |
| "learning_rate": 6.80037225320211e-05, |
| "loss": 0.1601, |
| "mean_token_accuracy": 0.925029194355011, |
| "num_tokens": 14979826.0, |
| "step": 160 |
| }, |
| { |
| "epoch": 0.12396694214876033, |
| "grad_norm": 0.230256125330925, |
| "learning_rate": 6.79088264941634e-05, |
| "loss": 0.1636, |
| "mean_token_accuracy": 0.9206058442592621, |
| "num_tokens": 15438466.0, |
| "step": 165 |
| }, |
| { |
| "epoch": 0.1277235161532682, |
| "grad_norm": 0.2395327240228653, |
| "learning_rate": 6.781030499225017e-05, |
| "loss": 0.1507, |
| "mean_token_accuracy": 0.9306641936302185, |
| "num_tokens": 15917450.0, |
| "step": 170 |
| }, |
| { |
| "epoch": 0.13148009015777612, |
| "grad_norm": 0.3818201422691345, |
| "learning_rate": 6.770817249917796e-05, |
| "loss": 0.144, |
| "mean_token_accuracy": 0.9311678409576416, |
| "num_tokens": 16392708.0, |
| "step": 175 |
| }, |
| { |
| "epoch": 0.135236664162284, |
| "grad_norm": 0.31338706612586975, |
| "learning_rate": 6.760244401830117e-05, |
| "loss": 0.1503, |
| "mean_token_accuracy": 0.9306272029876709, |
| "num_tokens": 16884068.0, |
| "step": 180 |
| }, |
| { |
| "epoch": 0.13899323816679188, |
| "grad_norm": 0.3805971145629883, |
| "learning_rate": 6.749313508122798e-05, |
| "loss": 0.1491, |
| "mean_token_accuracy": 0.9273220300674438, |
| "num_tokens": 17368574.0, |
| "step": 185 |
| }, |
| { |
| "epoch": 0.14274981217129978, |
| "grad_norm": 0.380010187625885, |
| "learning_rate": 6.738026174553883e-05, |
| "loss": 0.1562, |
| "mean_token_accuracy": 0.9291334331035614, |
| "num_tokens": 17836332.0, |
| "step": 190 |
| }, |
| { |
| "epoch": 0.14650638617580766, |
| "grad_norm": 0.32512524724006653, |
| "learning_rate": 6.726384059242745e-05, |
| "loss": 0.1531, |
| "mean_token_accuracy": 0.9281211197376251, |
| "num_tokens": 18296984.0, |
| "step": 195 |
| }, |
| { |
| "epoch": 0.15026296018031554, |
| "grad_norm": 0.20855890214443207, |
| "learning_rate": 6.714388872426516e-05, |
| "loss": 0.15, |
| "mean_token_accuracy": 0.9323076367378235, |
| "num_tokens": 18767919.0, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.15401953418482345, |
| "grad_norm": 0.6400463581085205, |
| "learning_rate": 6.702042376208847e-05, |
| "loss": 0.1459, |
| "mean_token_accuracy": 0.9299083292484284, |
| "num_tokens": 19223942.0, |
| "step": 205 |
| }, |
| { |
| "epoch": 0.15777610818933133, |
| "grad_norm": 0.34339508414268494, |
| "learning_rate": 6.689346384301047e-05, |
| "loss": 0.1471, |
| "mean_token_accuracy": 0.9301826536655426, |
| "num_tokens": 19688231.0, |
| "step": 210 |
| }, |
| { |
| "epoch": 0.1615326821938392, |
| "grad_norm": 0.41830775141716003, |
| "learning_rate": 6.676302761755661e-05, |
| "loss": 0.1602, |
| "mean_token_accuracy": 0.925974828004837, |
| "num_tokens": 20172877.0, |
| "step": 215 |
| }, |
| { |
| "epoch": 0.1652892561983471, |
| "grad_norm": 0.22839489579200745, |
| "learning_rate": 6.662913424692483e-05, |
| "loss": 0.1454, |
| "mean_token_accuracy": 0.9272423684597015, |
| "num_tokens": 20616865.0, |
| "step": 220 |
| }, |
| { |
| "epoch": 0.169045830202855, |
| "grad_norm": 0.4127933084964752, |
| "learning_rate": 6.649180340017077e-05, |
| "loss": 0.1415, |
| "mean_token_accuracy": 0.932004302740097, |
| "num_tokens": 21112172.0, |
| "step": 225 |
| }, |
| { |
| "epoch": 0.1728024042073629, |
| "grad_norm": 0.1962713748216629, |
| "learning_rate": 6.635105525131837e-05, |
| "loss": 0.1242, |
| "mean_token_accuracy": 0.9386635959148407, |
| "num_tokens": 21590051.0, |
| "step": 230 |
| }, |
| { |
| "epoch": 0.17655897821187078, |
| "grad_norm": 0.37941092252731323, |
| "learning_rate": 6.620691047639635e-05, |
| "loss": 0.1359, |
| "mean_token_accuracy": 0.9318796098232269, |
| "num_tokens": 22021981.0, |
| "step": 235 |
| }, |
| { |
| "epoch": 0.18031555221637866, |
| "grad_norm": 0.1975841522216797, |
| "learning_rate": 6.605939025040079e-05, |
| "loss": 0.1392, |
| "mean_token_accuracy": 0.932545930147171, |
| "num_tokens": 22493715.0, |
| "step": 240 |
| }, |
| { |
| "epoch": 0.18407212622088656, |
| "grad_norm": 0.26038625836372375, |
| "learning_rate": 6.590851624418453e-05, |
| "loss": 0.1559, |
| "mean_token_accuracy": 0.928155779838562, |
| "num_tokens": 22963724.0, |
| "step": 245 |
| }, |
| { |
| "epoch": 0.18782870022539444, |
| "grad_norm": 0.23218999803066254, |
| "learning_rate": 6.575431062127377e-05, |
| "loss": 0.1409, |
| "mean_token_accuracy": 0.9282140731811523, |
| "num_tokens": 23425234.0, |
| "step": 250 |
| }, |
| { |
| "epoch": 0.19158527422990232, |
| "grad_norm": 0.2822727859020233, |
| "learning_rate": 6.559679603461212e-05, |
| "loss": 0.1406, |
| "mean_token_accuracy": 0.9326273679733277, |
| "num_tokens": 23905900.0, |
| "step": 255 |
| }, |
| { |
| "epoch": 0.19534184823441023, |
| "grad_norm": 0.34615305066108704, |
| "learning_rate": 6.543599562323295e-05, |
| "loss": 0.1492, |
| "mean_token_accuracy": 0.925815999507904, |
| "num_tokens": 24354266.0, |
| "step": 260 |
| }, |
| { |
| "epoch": 0.1990984222389181, |
| "grad_norm": 0.3236545920372009, |
| "learning_rate": 6.527193300886021e-05, |
| "loss": 0.154, |
| "mean_token_accuracy": 0.9236413359642028, |
| "num_tokens": 24809551.0, |
| "step": 265 |
| }, |
| { |
| "epoch": 0.20285499624342598, |
| "grad_norm": 0.3167230188846588, |
| "learning_rate": 6.510463229243838e-05, |
| "loss": 0.1365, |
| "mean_token_accuracy": 0.9372122287750244, |
| "num_tokens": 25275924.0, |
| "step": 270 |
| }, |
| { |
| "epoch": 0.2066115702479339, |
| "grad_norm": 0.5786899924278259, |
| "learning_rate": 6.493411805059203e-05, |
| "loss": 0.1421, |
| "mean_token_accuracy": 0.9301037132740021, |
| "num_tokens": 25750072.0, |
| "step": 275 |
| } |
| ], |
| "logging_steps": 5, |
| "max_steps": 1331, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 1, |
| "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": 2.2411516276857897e+18, |
| "train_batch_size": 16, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
