| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.06765899864682003, | |
| "eval_steps": 9, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0013531799729364006, | |
| "grad_norm": 8.728355407714844, | |
| "learning_rate": 1e-05, | |
| "loss": 1.6889, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0013531799729364006, | |
| "eval_loss": 1.6367640495300293, | |
| "eval_runtime": 33.1413, | |
| "eval_samples_per_second": 18.798, | |
| "eval_steps_per_second": 2.354, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0027063599458728013, | |
| "grad_norm": 8.341392517089844, | |
| "learning_rate": 2e-05, | |
| "loss": 1.7952, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.0040595399188092015, | |
| "grad_norm": 8.935415267944336, | |
| "learning_rate": 3e-05, | |
| "loss": 1.5444, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.005412719891745603, | |
| "grad_norm": 7.361211776733398, | |
| "learning_rate": 4e-05, | |
| "loss": 1.452, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.006765899864682003, | |
| "grad_norm": 5.186268329620361, | |
| "learning_rate": 5e-05, | |
| "loss": 1.3568, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.008119079837618403, | |
| "grad_norm": 4.2360663414001465, | |
| "learning_rate": 6e-05, | |
| "loss": 1.1432, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.009472259810554804, | |
| "grad_norm": 3.4211342334747314, | |
| "learning_rate": 7e-05, | |
| "loss": 0.8908, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.010825439783491205, | |
| "grad_norm": 2.8270256519317627, | |
| "learning_rate": 8e-05, | |
| "loss": 0.7026, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.012178619756427604, | |
| "grad_norm": 4.626265525817871, | |
| "learning_rate": 9e-05, | |
| "loss": 0.7591, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.012178619756427604, | |
| "eval_loss": 0.4093446731567383, | |
| "eval_runtime": 32.5672, | |
| "eval_samples_per_second": 19.13, | |
| "eval_steps_per_second": 2.395, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.013531799729364006, | |
| "grad_norm": 1.9083399772644043, | |
| "learning_rate": 0.0001, | |
| "loss": 0.3278, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.014884979702300407, | |
| "grad_norm": 2.498229503631592, | |
| "learning_rate": 9.99695413509548e-05, | |
| "loss": 0.4199, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.016238159675236806, | |
| "grad_norm": 1.37482750415802, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 0.1413, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.017591339648173207, | |
| "grad_norm": 1.1291427612304688, | |
| "learning_rate": 9.972609476841367e-05, | |
| "loss": 0.137, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.018944519621109608, | |
| "grad_norm": 1.5424270629882812, | |
| "learning_rate": 9.951340343707852e-05, | |
| "loss": 0.1136, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.02029769959404601, | |
| "grad_norm": 1.4762475490570068, | |
| "learning_rate": 9.924038765061042e-05, | |
| "loss": 0.0587, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.02165087956698241, | |
| "grad_norm": 0.367654025554657, | |
| "learning_rate": 9.890738003669029e-05, | |
| "loss": 0.0216, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.023004059539918808, | |
| "grad_norm": 1.203076958656311, | |
| "learning_rate": 9.851478631379982e-05, | |
| "loss": 0.0476, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.02435723951285521, | |
| "grad_norm": 0.837530791759491, | |
| "learning_rate": 9.806308479691595e-05, | |
| "loss": 0.0208, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.02435723951285521, | |
| "eval_loss": 0.03585303574800491, | |
| "eval_runtime": 32.5664, | |
| "eval_samples_per_second": 19.13, | |
| "eval_steps_per_second": 2.395, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.02571041948579161, | |
| "grad_norm": 0.6422433853149414, | |
| "learning_rate": 9.755282581475769e-05, | |
| "loss": 0.0111, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.02706359945872801, | |
| "grad_norm": 1.1846771240234375, | |
| "learning_rate": 9.698463103929542e-05, | |
| "loss": 0.0873, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.028416779431664412, | |
| "grad_norm": 0.19324840605258942, | |
| "learning_rate": 9.635919272833938e-05, | |
| "loss": 0.0102, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.029769959404600813, | |
| "grad_norm": 0.296364426612854, | |
| "learning_rate": 9.567727288213005e-05, | |
| "loss": 0.0174, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.03112313937753721, | |
| "grad_norm": 0.7968310117721558, | |
| "learning_rate": 9.493970231495835e-05, | |
| "loss": 0.047, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.03247631935047361, | |
| "grad_norm": 0.12593711912631989, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 0.0061, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.03382949932341001, | |
| "grad_norm": 0.28613799810409546, | |
| "learning_rate": 9.330127018922194e-05, | |
| "loss": 0.0075, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.035182679296346414, | |
| "grad_norm": 0.23756200075149536, | |
| "learning_rate": 9.24024048078213e-05, | |
| "loss": 0.0165, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.036535859269282815, | |
| "grad_norm": 0.12324630469083786, | |
| "learning_rate": 9.145187862775209e-05, | |
| "loss": 0.0029, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.036535859269282815, | |
| "eval_loss": 0.020997991785407066, | |
| "eval_runtime": 32.6042, | |
| "eval_samples_per_second": 19.108, | |
| "eval_steps_per_second": 2.392, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.037889039242219216, | |
| "grad_norm": 0.08755990862846375, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 0.0074, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.03924221921515562, | |
| "grad_norm": 0.18577858805656433, | |
| "learning_rate": 8.940053768033609e-05, | |
| "loss": 0.0084, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.04059539918809202, | |
| "grad_norm": 0.09318247437477112, | |
| "learning_rate": 8.83022221559489e-05, | |
| "loss": 0.0049, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.04194857916102842, | |
| "grad_norm": 0.2695228159427643, | |
| "learning_rate": 8.715724127386972e-05, | |
| "loss": 0.0159, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.04330175913396482, | |
| "grad_norm": 0.08464854955673218, | |
| "learning_rate": 8.596699001693255e-05, | |
| "loss": 0.0065, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.044654939106901215, | |
| "grad_norm": 0.17591221630573273, | |
| "learning_rate": 8.473291852294987e-05, | |
| "loss": 0.0049, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.046008119079837616, | |
| "grad_norm": 0.2792809009552002, | |
| "learning_rate": 8.345653031794292e-05, | |
| "loss": 0.0171, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.04736129905277402, | |
| "grad_norm": 0.12685370445251465, | |
| "learning_rate": 8.213938048432697e-05, | |
| "loss": 0.0043, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.04871447902571042, | |
| "grad_norm": 0.5813450813293457, | |
| "learning_rate": 8.07830737662829e-05, | |
| "loss": 0.0135, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.04871447902571042, | |
| "eval_loss": 0.019910942763090134, | |
| "eval_runtime": 32.5995, | |
| "eval_samples_per_second": 19.111, | |
| "eval_steps_per_second": 2.393, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.05006765899864682, | |
| "grad_norm": 0.08670137077569962, | |
| "learning_rate": 7.938926261462366e-05, | |
| "loss": 0.0062, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.05142083897158322, | |
| "grad_norm": 0.09463904052972794, | |
| "learning_rate": 7.795964517353735e-05, | |
| "loss": 0.0021, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.05277401894451962, | |
| "grad_norm": 0.05870044231414795, | |
| "learning_rate": 7.649596321166024e-05, | |
| "loss": 0.006, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.05412719891745602, | |
| "grad_norm": 0.09653154015541077, | |
| "learning_rate": 7.500000000000001e-05, | |
| "loss": 0.007, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.05548037889039242, | |
| "grad_norm": 0.06797784566879272, | |
| "learning_rate": 7.347357813929454e-05, | |
| "loss": 0.0094, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.056833558863328824, | |
| "grad_norm": 0.9252069592475891, | |
| "learning_rate": 7.191855733945387e-05, | |
| "loss": 0.0755, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.058186738836265225, | |
| "grad_norm": 0.08743719756603241, | |
| "learning_rate": 7.033683215379002e-05, | |
| "loss": 0.007, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.05953991880920163, | |
| "grad_norm": 0.22028687596321106, | |
| "learning_rate": 6.873032967079561e-05, | |
| "loss": 0.0107, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.06089309878213803, | |
| "grad_norm": 0.24319298565387726, | |
| "learning_rate": 6.710100716628344e-05, | |
| "loss": 0.0168, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.06089309878213803, | |
| "eval_loss": 0.019319333136081696, | |
| "eval_runtime": 32.603, | |
| "eval_samples_per_second": 19.109, | |
| "eval_steps_per_second": 2.392, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.06224627875507442, | |
| "grad_norm": 0.6694943904876709, | |
| "learning_rate": 6.545084971874738e-05, | |
| "loss": 0.081, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.06359945872801083, | |
| "grad_norm": 0.06383314728736877, | |
| "learning_rate": 6.378186779084995e-05, | |
| "loss": 0.0016, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.06495263870094722, | |
| "grad_norm": 0.05941140279173851, | |
| "learning_rate": 6.209609477998338e-05, | |
| "loss": 0.0045, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.06630581867388363, | |
| "grad_norm": 0.19987758994102478, | |
| "learning_rate": 6.0395584540887963e-05, | |
| "loss": 0.0132, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.06765899864682003, | |
| "grad_norm": 0.11729590594768524, | |
| "learning_rate": 5.868240888334653e-05, | |
| "loss": 0.0058, | |
| "step": 50 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 100, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 25, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 3.70943641780224e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |