| { |
| "best_global_step": null, |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 1.0, |
| "eval_steps": 500, |
| "global_step": 740, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "entropy": 1.810372607409954, |
| "epoch": 0.033818058843422386, |
| "grad_norm": 0.03564453125, |
| "learning_rate": 0.00019999904008949706, |
| "loss": 1.961472625732422, |
| "mean_token_accuracy": 0.5825667985156179, |
| "num_tokens": 156162.0, |
| "step": 25 |
| }, |
| { |
| "entropy": 1.630317051112652, |
| "epoch": 0.06763611768684477, |
| "grad_norm": 0.06396484375, |
| "learning_rate": 0.00019935180094484163, |
| "loss": 1.5934637451171876, |
| "mean_token_accuracy": 0.6405555133521557, |
| "num_tokens": 314902.0, |
| "step": 50 |
| }, |
| { |
| "entropy": 1.4665888720750808, |
| "epoch": 0.10145417653026716, |
| "grad_norm": 0.1220703125, |
| "learning_rate": 0.00019751364095427692, |
| "loss": 1.430908203125, |
| "mean_token_accuracy": 0.6695231589674949, |
| "num_tokens": 466632.0, |
| "step": 75 |
| }, |
| { |
| "entropy": 1.392833354473114, |
| "epoch": 0.13527223537368954, |
| "grad_norm": 0.134765625, |
| "learning_rate": 0.00019450659397154355, |
| "loss": 1.3299276733398437, |
| "mean_token_accuracy": 0.6912748428061605, |
| "num_tokens": 624850.0, |
| "step": 100 |
| }, |
| { |
| "entropy": 1.1405542536824942, |
| "epoch": 0.16909029421711194, |
| "grad_norm": 0.1318359375, |
| "learning_rate": 0.00019036670519061188, |
| "loss": 1.0945913696289062, |
| "mean_token_accuracy": 0.7419995337724685, |
| "num_tokens": 781370.0, |
| "step": 125 |
| }, |
| { |
| "entropy": 1.0725987593457103, |
| "epoch": 0.2029083530605343, |
| "grad_norm": 0.09619140625, |
| "learning_rate": 0.0001851435990752769, |
| "loss": 1.0093214416503906, |
| "mean_token_accuracy": 0.7651712036132813, |
| "num_tokens": 938626.0, |
| "step": 150 |
| }, |
| { |
| "entropy": 0.9529129752889276, |
| "epoch": 0.2367264119039567, |
| "grad_norm": 0.08349609375, |
| "learning_rate": 0.0001788998845153021, |
| "loss": 0.8964275360107422, |
| "mean_token_accuracy": 0.7902284222841263, |
| "num_tokens": 1094352.0, |
| "step": 175 |
| }, |
| { |
| "entropy": 0.862411651480943, |
| "epoch": 0.2705444707473791, |
| "grad_norm": 0.09033203125, |
| "learning_rate": 0.00017171040433945005, |
| "loss": 0.8004817199707032, |
| "mean_token_accuracy": 0.8113848954439163, |
| "num_tokens": 1251490.0, |
| "step": 200 |
| }, |
| { |
| "entropy": 0.7631507751531899, |
| "epoch": 0.3043625295908015, |
| "grad_norm": 0.09228515625, |
| "learning_rate": 0.00016366133818141893, |
| "loss": 0.7039064025878906, |
| "mean_token_accuracy": 0.8357413372397423, |
| "num_tokens": 1406770.0, |
| "step": 225 |
| }, |
| { |
| "entropy": 0.7098334492836148, |
| "epoch": 0.3381805884342239, |
| "grad_norm": 0.0732421875, |
| "learning_rate": 0.0001548491694525464, |
| "loss": 0.6646959686279297, |
| "mean_token_accuracy": 0.8501214118301869, |
| "num_tokens": 1561213.0, |
| "step": 250 |
| }, |
| { |
| "entropy": 0.7675167324021459, |
| "epoch": 0.37199864727764625, |
| "grad_norm": 0.1044921875, |
| "learning_rate": 0.00014537952880408473, |
| "loss": 0.71765869140625, |
| "mean_token_accuracy": 0.835702428817749, |
| "num_tokens": 1724371.0, |
| "step": 275 |
| }, |
| { |
| "entropy": 0.6311481145210565, |
| "epoch": 0.4058167061210686, |
| "grad_norm": 0.08203125, |
| "learning_rate": 0.00013536592794235696, |
| "loss": 0.5955865478515625, |
| "mean_token_accuracy": 0.8682514359802007, |
| "num_tokens": 1885062.0, |
| "step": 300 |
| }, |
| { |
| "entropy": 0.7570231725927442, |
| "epoch": 0.43963476496449105, |
| "grad_norm": 0.10693359375, |
| "learning_rate": 0.00012492839897443553, |
| "loss": 0.7146268463134766, |
| "mean_token_accuracy": 0.8396708873659372, |
| "num_tokens": 2043297.0, |
| "step": 325 |
| }, |
| { |
| "entropy": 0.5249405995476991, |
| "epoch": 0.4734528238079134, |
| "grad_norm": 0.078125, |
| "learning_rate": 0.00011419205559437998, |
| "loss": 0.49004329681396486, |
| "mean_token_accuracy": 0.8892295673489571, |
| "num_tokens": 2196510.0, |
| "step": 350 |
| }, |
| { |
| "entropy": 0.6064728916622698, |
| "epoch": 0.5072708826513358, |
| "grad_norm": 0.07470703125, |
| "learning_rate": 0.00010328559335696188, |
| "loss": 0.5641261291503906, |
| "mean_token_accuracy": 0.8721542170643807, |
| "num_tokens": 2354986.0, |
| "step": 375 |
| }, |
| { |
| "entropy": 0.5784782155323773, |
| "epoch": 0.5410889414947582, |
| "grad_norm": 0.07958984375, |
| "learning_rate": 9.233974701595655e-05, |
| "loss": 0.5496908950805665, |
| "mean_token_accuracy": 0.8779391979053617, |
| "num_tokens": 2515773.0, |
| "step": 400 |
| }, |
| { |
| "entropy": 0.5494727499317378, |
| "epoch": 0.5749070003381805, |
| "grad_norm": 0.053466796875, |
| "learning_rate": 8.14857234187453e-05, |
| "loss": 0.5182221984863281, |
| "mean_token_accuracy": 0.8837397874891758, |
| "num_tokens": 2674952.0, |
| "step": 425 |
| }, |
| { |
| "entropy": 0.5801290932204575, |
| "epoch": 0.608725059181603, |
| "grad_norm": 0.091796875, |
| "learning_rate": 7.08536287419749e-05, |
| "loss": 0.5686647033691407, |
| "mean_token_accuracy": 0.8799975875392556, |
| "num_tokens": 2830234.0, |
| "step": 450 |
| }, |
| { |
| "entropy": 0.48558603684417906, |
| "epoch": 0.6425431180250254, |
| "grad_norm": 0.050537109375, |
| "learning_rate": 6.0570908920855706e-05, |
| "loss": 0.42860363006591795, |
| "mean_token_accuracy": 0.9008280988037586, |
| "num_tokens": 2989837.0, |
| "step": 475 |
| }, |
| { |
| "entropy": 0.4568167580384761, |
| "epoch": 0.6763611768684478, |
| "grad_norm": 0.095703125, |
| "learning_rate": 5.076082196652754e-05, |
| "loss": 0.45085529327392576, |
| "mean_token_accuracy": 0.9038965445756912, |
| "num_tokens": 3143012.0, |
| "step": 500 |
| }, |
| { |
| "entropy": 0.44007344283163546, |
| "epoch": 0.7101792357118701, |
| "grad_norm": 0.058837890625, |
| "learning_rate": 4.15409604836838e-05, |
| "loss": 0.41196117401123045, |
| "mean_token_accuracy": 0.9095691731572151, |
| "num_tokens": 3300477.0, |
| "step": 525 |
| }, |
| { |
| "entropy": 0.4962937253713608, |
| "epoch": 0.7439972945552925, |
| "grad_norm": 0.0537109375, |
| "learning_rate": 3.3021842098898934e-05, |
| "loss": 0.4741514205932617, |
| "mean_token_accuracy": 0.896233549118042, |
| "num_tokens": 3454477.0, |
| "step": 550 |
| }, |
| { |
| "entropy": 0.45510641247965394, |
| "epoch": 0.7778153533987149, |
| "grad_norm": 0.06298828125, |
| "learning_rate": 2.530558469606351e-05, |
| "loss": 0.4256900787353516, |
| "mean_token_accuracy": 0.9090374359488487, |
| "num_tokens": 3614381.0, |
| "step": 575 |
| }, |
| { |
| "entropy": 0.4509261836577207, |
| "epoch": 0.8116334122421373, |
| "grad_norm": 0.060791015625, |
| "learning_rate": 1.8484682338758152e-05, |
| "loss": 0.4140806198120117, |
| "mean_token_accuracy": 0.9066468042135238, |
| "num_tokens": 3770302.0, |
| "step": 600 |
| }, |
| { |
| "entropy": 0.48186022156849506, |
| "epoch": 0.8454514710855597, |
| "grad_norm": 0.060302734375, |
| "learning_rate": 1.2640896552470794e-05, |
| "loss": 0.4636331939697266, |
| "mean_token_accuracy": 0.9016842505335808, |
| "num_tokens": 3929478.0, |
| "step": 625 |
| }, |
| { |
| "entropy": 0.45708779353648427, |
| "epoch": 0.8792695299289821, |
| "grad_norm": 0.0125732421875, |
| "learning_rate": 7.844276256755346e-06, |
| "loss": 0.45078208923339846, |
| "mean_token_accuracy": 0.9030292834341526, |
| "num_tokens": 4084974.0, |
| "step": 650 |
| }, |
| { |
| "entropy": 0.5012604260630905, |
| "epoch": 0.9130875887724045, |
| "grad_norm": 0.072265625, |
| "learning_rate": 4.152318095313778e-06, |
| "loss": 0.47621334075927735, |
| "mean_token_accuracy": 0.8943454670906067, |
| "num_tokens": 4246619.0, |
| "step": 675 |
| }, |
| { |
| "entropy": 0.41868842981755733, |
| "epoch": 0.9469056476158269, |
| "grad_norm": 0.0341796875, |
| "learning_rate": 1.609277229047801e-06, |
| "loss": 0.38456172943115235, |
| "mean_token_accuracy": 0.9145496153831482, |
| "num_tokens": 4405485.0, |
| "step": 700 |
| }, |
| { |
| "entropy": 0.4935927293263376, |
| "epoch": 0.9807237064592492, |
| "grad_norm": 0.0732421875, |
| "learning_rate": 2.4563685353949925e-07, |
| "loss": 0.47506561279296877, |
| "mean_token_accuracy": 0.8976721769571304, |
| "num_tokens": 4564678.0, |
| "step": 725 |
| } |
| ], |
| "logging_steps": 25, |
| "max_steps": 740, |
| "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": true |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 1.9858744653595546e+17, |
| "train_batch_size": 1, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
