| { |
| "best_global_step": null, |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 1.0, |
| "eval_steps": 500, |
| "global_step": 325, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "ce_loss": 0.08120492927497253, |
| "epoch": 0.03076923076923077, |
| "grad_norm": 2.71875, |
| "kl_loss": 0.25050720800128246, |
| "learning_rate": 4.5e-05, |
| "loss": 0.6271, |
| "mean_token_accuracy": 0.6869053975741068, |
| "num_tokens": 32469.0, |
| "step": 10 |
| }, |
| { |
| "ce_loss": 0.06159211456232394, |
| "epoch": 0.06153846153846154, |
| "grad_norm": 1.5, |
| "kl_loss": 0.25381727303777424, |
| "learning_rate": 4.989935734988098e-05, |
| "loss": 0.4815, |
| "mean_token_accuracy": 0.7235488637288412, |
| "num_tokens": 58642.0, |
| "step": 20 |
| }, |
| { |
| "ce_loss": 0.07180731555108184, |
| "epoch": 0.09230769230769231, |
| "grad_norm": 2.03125, |
| "kl_loss": 0.24764725762522882, |
| "learning_rate": 4.9552497026005974e-05, |
| "loss": 0.5779, |
| "mean_token_accuracy": 0.7307767424827967, |
| "num_tokens": 93193.0, |
| "step": 30 |
| }, |
| { |
| "ce_loss": 0.058572557456972825, |
| "epoch": 0.12307692307692308, |
| "grad_norm": 1.328125, |
| "kl_loss": 0.24026942600806553, |
| "learning_rate": 4.896162295600589e-05, |
| "loss": 0.469, |
| "mean_token_accuracy": 0.7393925120304157, |
| "num_tokens": 121044.0, |
| "step": 40 |
| }, |
| { |
| "ce_loss": 0.0657964160782285, |
| "epoch": 0.15384615384615385, |
| "grad_norm": 1.484375, |
| "kl_loss": 0.23294192283517784, |
| "learning_rate": 4.813260751184992e-05, |
| "loss": 0.5168, |
| "mean_token_accuracy": 0.7145701113500094, |
| "num_tokens": 153295.0, |
| "step": 50 |
| }, |
| { |
| "ce_loss": 0.06406597602513751, |
| "epoch": 0.18461538461538463, |
| "grad_norm": 1.4375, |
| "kl_loss": 0.2228865826295482, |
| "learning_rate": 4.707368982147318e-05, |
| "loss": 0.4645, |
| "mean_token_accuracy": 0.7282789766788482, |
| "num_tokens": 180604.0, |
| "step": 60 |
| }, |
| { |
| "ce_loss": 0.061541669910798805, |
| "epoch": 0.2153846153846154, |
| "grad_norm": 1.8046875, |
| "kl_loss": 0.23179643671657588, |
| "learning_rate": 4.579539388462173e-05, |
| "loss": 0.4726, |
| "mean_token_accuracy": 0.7627677949699195, |
| "num_tokens": 209749.0, |
| "step": 70 |
| }, |
| { |
| "ce_loss": 0.06250920130828903, |
| "epoch": 0.24615384615384617, |
| "grad_norm": 1.453125, |
| "kl_loss": 0.2295288745459024, |
| "learning_rate": 4.431042398061499e-05, |
| "loss": 0.499, |
| "mean_token_accuracy": 0.7377513333574518, |
| "num_tokens": 241565.0, |
| "step": 80 |
| }, |
| { |
| "ce_loss": 0.06265452711454902, |
| "epoch": 0.27692307692307694, |
| "grad_norm": 1.5078125, |
| "kl_loss": 0.21842536350635633, |
| "learning_rate": 4.263353840751022e-05, |
| "loss": 0.4859, |
| "mean_token_accuracy": 0.7372650110721588, |
| "num_tokens": 270576.0, |
| "step": 90 |
| }, |
| { |
| "ce_loss": 0.07052235386436223, |
| "epoch": 0.3076923076923077, |
| "grad_norm": 1.4765625, |
| "kl_loss": 0.23768770654996235, |
| "learning_rate": 4.078140280750597e-05, |
| "loss": 0.5256, |
| "mean_token_accuracy": 0.7432598421971003, |
| "num_tokens": 304058.0, |
| "step": 100 |
| }, |
| { |
| "ce_loss": 0.0685534270608817, |
| "epoch": 0.3384615384615385, |
| "grad_norm": 1.453125, |
| "kl_loss": 0.20998213719576597, |
| "learning_rate": 3.8772424536302564e-05, |
| "loss": 0.5087, |
| "mean_token_accuracy": 0.7263170762194527, |
| "num_tokens": 333547.0, |
| "step": 110 |
| }, |
| { |
| "ce_loss": 0.061361576650971035, |
| "epoch": 0.36923076923076925, |
| "grad_norm": 1.328125, |
| "kl_loss": 0.2288582260279279, |
| "learning_rate": 3.662656972253127e-05, |
| "loss": 0.4653, |
| "mean_token_accuracy": 0.7384924937600958, |
| "num_tokens": 363397.0, |
| "step": 120 |
| }, |
| { |
| "ce_loss": 0.058161119158845795, |
| "epoch": 0.4, |
| "grad_norm": 1.546875, |
| "kl_loss": 0.22008681535720825, |
| "learning_rate": 3.436516483539781e-05, |
| "loss": 0.4411, |
| "mean_token_accuracy": 0.7447177498307946, |
| "num_tokens": 390870.0, |
| "step": 130 |
| }, |
| { |
| "ce_loss": 0.06444319941893857, |
| "epoch": 0.4307692307692308, |
| "grad_norm": 1.5078125, |
| "kl_loss": 0.22319350773707414, |
| "learning_rate": 3.201068473265007e-05, |
| "loss": 0.5072, |
| "mean_token_accuracy": 0.7319575532486564, |
| "num_tokens": 421547.0, |
| "step": 140 |
| }, |
| { |
| "ce_loss": 0.07355413343054364, |
| "epoch": 0.46153846153846156, |
| "grad_norm": 2.25, |
| "kl_loss": 0.22494713615064751, |
| "learning_rate": 2.958652929534456e-05, |
| "loss": 0.5828, |
| "mean_token_accuracy": 0.7224259686160397, |
| "num_tokens": 455995.0, |
| "step": 150 |
| }, |
| { |
| "ce_loss": 0.0639672374632533, |
| "epoch": 0.49230769230769234, |
| "grad_norm": 1.28125, |
| "kl_loss": 0.22526652898107255, |
| "learning_rate": 2.7116790869315582e-05, |
| "loss": 0.4763, |
| "mean_token_accuracy": 0.7467372723751597, |
| "num_tokens": 485763.0, |
| "step": 160 |
| }, |
| { |
| "ce_loss": 0.06010204482696612, |
| "epoch": 0.5230769230769231, |
| "grad_norm": 1.640625, |
| "kl_loss": 0.21317672141288457, |
| "learning_rate": 2.4626014824618415e-05, |
| "loss": 0.479, |
| "mean_token_accuracy": 0.7542458176612854, |
| "num_tokens": 517224.0, |
| "step": 170 |
| }, |
| { |
| "ce_loss": 0.06867885408133294, |
| "epoch": 0.5538461538461539, |
| "grad_norm": 1.796875, |
| "kl_loss": 0.21281489445582816, |
| "learning_rate": 2.2138955612614207e-05, |
| "loss": 0.5244, |
| "mean_token_accuracy": 0.7292757477309253, |
| "num_tokens": 546798.0, |
| "step": 180 |
| }, |
| { |
| "ce_loss": 0.07290608537311384, |
| "epoch": 0.5846153846153846, |
| "grad_norm": 1.71875, |
| "kl_loss": 0.21553313066562016, |
| "learning_rate": 1.9680330745110954e-05, |
| "loss": 0.5703, |
| "mean_token_accuracy": 0.7481181339213723, |
| "num_tokens": 584214.0, |
| "step": 190 |
| }, |
| { |
| "ce_loss": 0.06328318048197823, |
| "epoch": 0.6153846153846154, |
| "grad_norm": 1.921875, |
| "kl_loss": 0.21544352812426432, |
| "learning_rate": 1.7274575140626318e-05, |
| "loss": 0.5165, |
| "mean_token_accuracy": 0.738218150561369, |
| "num_tokens": 615971.0, |
| "step": 200 |
| }, |
| { |
| "ce_loss": 0.07065691198828336, |
| "epoch": 0.6461538461538462, |
| "grad_norm": 1.8125, |
| "kl_loss": 0.21665606206991303, |
| "learning_rate": 1.4945598279189565e-05, |
| "loss": 0.5312, |
| "mean_token_accuracy": 0.7369330035497065, |
| "num_tokens": 647849.0, |
| "step": 210 |
| }, |
| { |
| "ce_loss": 0.06142696152084398, |
| "epoch": 0.676923076923077, |
| "grad_norm": 2.015625, |
| "kl_loss": 0.2208457773923874, |
| "learning_rate": 1.271654657918722e-05, |
| "loss": 0.4896, |
| "mean_token_accuracy": 0.73205061314942, |
| "num_tokens": 678046.0, |
| "step": 220 |
| }, |
| { |
| "ce_loss": 0.059431257970672585, |
| "epoch": 0.7076923076923077, |
| "grad_norm": 1.71875, |
| "kl_loss": 0.22038192882553323, |
| "learning_rate": 1.0609573357858166e-05, |
| "loss": 0.4508, |
| "mean_token_accuracy": 0.739933922682723, |
| "num_tokens": 704690.0, |
| "step": 230 |
| }, |
| { |
| "ce_loss": 0.06104397588373663, |
| "epoch": 0.7384615384615385, |
| "grad_norm": 1.609375, |
| "kl_loss": 0.20902339330354294, |
| "learning_rate": 8.645618661674142e-06, |
| "loss": 0.4922, |
| "mean_token_accuracy": 0.7469771626668099, |
| "num_tokens": 736788.0, |
| "step": 240 |
| }, |
| { |
| "ce_loss": 0.07018770978825878, |
| "epoch": 0.7692307692307693, |
| "grad_norm": 1.578125, |
| "kl_loss": 0.2098122204483395, |
| "learning_rate": 6.844201154750177e-06, |
| "loss": 0.5483, |
| "mean_token_accuracy": 0.7493137827045039, |
| "num_tokens": 774271.0, |
| "step": 250 |
| }, |
| { |
| "ce_loss": 0.06064042067931344, |
| "epoch": 0.8, |
| "grad_norm": 1.8046875, |
| "kl_loss": 0.21842047728990255, |
| "learning_rate": 5.223224133591476e-06, |
| "loss": 0.4714, |
| "mean_token_accuracy": 0.7411498236656189, |
| "num_tokens": 801718.0, |
| "step": 260 |
| }, |
| { |
| "ce_loss": 0.05912412864307879, |
| "epoch": 0.8307692307692308, |
| "grad_norm": 1.46875, |
| "kl_loss": 0.22446780859843476, |
| "learning_rate": 3.798797596089351e-06, |
| "loss": 0.4548, |
| "mean_token_accuracy": 0.7460238772469598, |
| "num_tokens": 829369.0, |
| "step": 270 |
| }, |
| { |
| "ce_loss": 0.06512451142847758, |
| "epoch": 0.8615384615384616, |
| "grad_norm": 1.7265625, |
| "kl_loss": 0.20963396688302358, |
| "learning_rate": 2.58507813312448e-06, |
| "loss": 0.5107, |
| "mean_token_accuracy": 0.7250431329011917, |
| "num_tokens": 857145.0, |
| "step": 280 |
| }, |
| { |
| "ce_loss": 0.06407053675664226, |
| "epoch": 0.8923076923076924, |
| "grad_norm": 1.171875, |
| "kl_loss": 0.2203041362135034, |
| "learning_rate": 1.59412823400657e-06, |
| "loss": 0.4781, |
| "mean_token_accuracy": 0.7484459264410867, |
| "num_tokens": 887311.0, |
| "step": 290 |
| }, |
| { |
| "ce_loss": 0.0664482081074557, |
| "epoch": 0.9230769230769231, |
| "grad_norm": 1.484375, |
| "kl_loss": 0.21361721268831155, |
| "learning_rate": 8.357964040363209e-07, |
| "loss": 0.5084, |
| "mean_token_accuracy": 0.7330168260110391, |
| "num_tokens": 918618.0, |
| "step": 300 |
| }, |
| { |
| "ce_loss": 0.07091758686992516, |
| "epoch": 0.9538461538461539, |
| "grad_norm": 1.3046875, |
| "kl_loss": 0.22180311183134716, |
| "learning_rate": 3.1761928563510955e-07, |
| "loss": 0.5414, |
| "mean_token_accuracy": 0.7199107682382738, |
| "num_tokens": 952762.0, |
| "step": 310 |
| }, |
| { |
| "ce_loss": 0.05512376611185984, |
| "epoch": 0.9846153846153847, |
| "grad_norm": 1.328125, |
| "kl_loss": 0.21334302270567262, |
| "learning_rate": 4.474675580662113e-08, |
| "loss": 0.4188, |
| "mean_token_accuracy": 0.7297707796096802, |
| "num_tokens": 979999.0, |
| "step": 320 |
| } |
| ], |
| "logging_steps": 10, |
| "max_steps": 325, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 1, |
| "save_steps": -325, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": true |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 1.427496913409065e+17, |
| "train_batch_size": 4, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
