| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.3968253968253968, | |
| "eval_steps": 9, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.007936507936507936, | |
| "grad_norm": 0.5940561294555664, | |
| "learning_rate": 1e-05, | |
| "loss": 0.7819, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.007936507936507936, | |
| "eval_loss": 0.7385377883911133, | |
| "eval_runtime": 12.4705, | |
| "eval_samples_per_second": 8.5, | |
| "eval_steps_per_second": 1.123, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.015873015873015872, | |
| "grad_norm": 0.5844579339027405, | |
| "learning_rate": 2e-05, | |
| "loss": 0.5743, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.023809523809523808, | |
| "grad_norm": 0.6142908334732056, | |
| "learning_rate": 3e-05, | |
| "loss": 0.7856, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.031746031746031744, | |
| "grad_norm": 1.1579065322875977, | |
| "learning_rate": 4e-05, | |
| "loss": 1.0332, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.03968253968253968, | |
| "grad_norm": 0.6027746796607971, | |
| "learning_rate": 5e-05, | |
| "loss": 0.6717, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.047619047619047616, | |
| "grad_norm": 0.9417018294334412, | |
| "learning_rate": 6e-05, | |
| "loss": 0.7161, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.05555555555555555, | |
| "grad_norm": 0.8347136974334717, | |
| "learning_rate": 7e-05, | |
| "loss": 0.7583, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.06349206349206349, | |
| "grad_norm": 1.023072361946106, | |
| "learning_rate": 8e-05, | |
| "loss": 0.5415, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.07142857142857142, | |
| "grad_norm": 0.4929274022579193, | |
| "learning_rate": 9e-05, | |
| "loss": 0.4157, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.07142857142857142, | |
| "eval_loss": 0.4522353708744049, | |
| "eval_runtime": 11.789, | |
| "eval_samples_per_second": 8.991, | |
| "eval_steps_per_second": 1.188, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.07936507936507936, | |
| "grad_norm": 0.5577216148376465, | |
| "learning_rate": 0.0001, | |
| "loss": 0.4259, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.0873015873015873, | |
| "grad_norm": 0.7210971117019653, | |
| "learning_rate": 9.99695413509548e-05, | |
| "loss": 0.4377, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.09523809523809523, | |
| "grad_norm": 0.4076630473136902, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 0.3398, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.10317460317460317, | |
| "grad_norm": 0.6704097986221313, | |
| "learning_rate": 9.972609476841367e-05, | |
| "loss": 0.3923, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.1111111111111111, | |
| "grad_norm": 0.6116910576820374, | |
| "learning_rate": 9.951340343707852e-05, | |
| "loss": 0.2966, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.11904761904761904, | |
| "grad_norm": 0.9248013496398926, | |
| "learning_rate": 9.924038765061042e-05, | |
| "loss": 0.3506, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.12698412698412698, | |
| "grad_norm": 0.7895079851150513, | |
| "learning_rate": 9.890738003669029e-05, | |
| "loss": 0.2511, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.1349206349206349, | |
| "grad_norm": 0.4172516167163849, | |
| "learning_rate": 9.851478631379982e-05, | |
| "loss": 0.2474, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.14285714285714285, | |
| "grad_norm": 0.4765908718109131, | |
| "learning_rate": 9.806308479691595e-05, | |
| "loss": 0.2699, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.14285714285714285, | |
| "eval_loss": 0.2643006443977356, | |
| "eval_runtime": 11.7881, | |
| "eval_samples_per_second": 8.992, | |
| "eval_steps_per_second": 1.188, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.15079365079365079, | |
| "grad_norm": 0.5411407947540283, | |
| "learning_rate": 9.755282581475769e-05, | |
| "loss": 0.2662, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.15873015873015872, | |
| "grad_norm": 0.3574010729789734, | |
| "learning_rate": 9.698463103929542e-05, | |
| "loss": 0.2805, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.16666666666666666, | |
| "grad_norm": 0.7612572312355042, | |
| "learning_rate": 9.635919272833938e-05, | |
| "loss": 0.3236, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.1746031746031746, | |
| "grad_norm": 0.3988335132598877, | |
| "learning_rate": 9.567727288213005e-05, | |
| "loss": 0.2418, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.18253968253968253, | |
| "grad_norm": 0.39348870515823364, | |
| "learning_rate": 9.493970231495835e-05, | |
| "loss": 0.2039, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.19047619047619047, | |
| "grad_norm": 0.6038708090782166, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 0.2575, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.1984126984126984, | |
| "grad_norm": 0.6375281810760498, | |
| "learning_rate": 9.330127018922194e-05, | |
| "loss": 0.2512, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.20634920634920634, | |
| "grad_norm": 0.482608437538147, | |
| "learning_rate": 9.24024048078213e-05, | |
| "loss": 0.2392, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.21428571428571427, | |
| "grad_norm": 0.35353580117225647, | |
| "learning_rate": 9.145187862775209e-05, | |
| "loss": 0.232, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.21428571428571427, | |
| "eval_loss": 0.23887650668621063, | |
| "eval_runtime": 11.7951, | |
| "eval_samples_per_second": 8.987, | |
| "eval_steps_per_second": 1.187, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.2222222222222222, | |
| "grad_norm": 0.5033824443817139, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 0.2169, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.23015873015873015, | |
| "grad_norm": 0.4087642431259155, | |
| "learning_rate": 8.940053768033609e-05, | |
| "loss": 0.2463, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.23809523809523808, | |
| "grad_norm": 0.3605215549468994, | |
| "learning_rate": 8.83022221559489e-05, | |
| "loss": 0.2425, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.24603174603174602, | |
| "grad_norm": 0.2426234781742096, | |
| "learning_rate": 8.715724127386972e-05, | |
| "loss": 0.2237, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.25396825396825395, | |
| "grad_norm": 0.4276462197303772, | |
| "learning_rate": 8.596699001693255e-05, | |
| "loss": 0.263, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.2619047619047619, | |
| "grad_norm": 0.3156845271587372, | |
| "learning_rate": 8.473291852294987e-05, | |
| "loss": 0.2098, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.2698412698412698, | |
| "grad_norm": 0.22470353543758392, | |
| "learning_rate": 8.345653031794292e-05, | |
| "loss": 0.1993, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.2777777777777778, | |
| "grad_norm": 0.23595388233661652, | |
| "learning_rate": 8.213938048432697e-05, | |
| "loss": 0.2161, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.2857142857142857, | |
| "grad_norm": 0.3054620325565338, | |
| "learning_rate": 8.07830737662829e-05, | |
| "loss": 0.2123, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.2857142857142857, | |
| "eval_loss": 0.22184596955776215, | |
| "eval_runtime": 11.7876, | |
| "eval_samples_per_second": 8.993, | |
| "eval_steps_per_second": 1.188, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.29365079365079366, | |
| "grad_norm": 0.42485561966896057, | |
| "learning_rate": 7.938926261462366e-05, | |
| "loss": 0.2481, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.30158730158730157, | |
| "grad_norm": 0.25488778948783875, | |
| "learning_rate": 7.795964517353735e-05, | |
| "loss": 0.2089, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.30952380952380953, | |
| "grad_norm": 0.22732485830783844, | |
| "learning_rate": 7.649596321166024e-05, | |
| "loss": 0.2151, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.31746031746031744, | |
| "grad_norm": 0.569657027721405, | |
| "learning_rate": 7.500000000000001e-05, | |
| "loss": 0.1922, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.3253968253968254, | |
| "grad_norm": 0.27273625135421753, | |
| "learning_rate": 7.347357813929454e-05, | |
| "loss": 0.2203, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.3333333333333333, | |
| "grad_norm": 0.547612726688385, | |
| "learning_rate": 7.191855733945387e-05, | |
| "loss": 0.2446, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.3412698412698413, | |
| "grad_norm": 0.3911803364753723, | |
| "learning_rate": 7.033683215379002e-05, | |
| "loss": 0.2097, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.3492063492063492, | |
| "grad_norm": 0.34991058707237244, | |
| "learning_rate": 6.873032967079561e-05, | |
| "loss": 0.1864, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.35714285714285715, | |
| "grad_norm": 0.41730400919914246, | |
| "learning_rate": 6.710100716628344e-05, | |
| "loss": 0.2209, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.35714285714285715, | |
| "eval_loss": 0.2063593566417694, | |
| "eval_runtime": 11.7486, | |
| "eval_samples_per_second": 9.022, | |
| "eval_steps_per_second": 1.192, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.36507936507936506, | |
| "grad_norm": 0.48006361722946167, | |
| "learning_rate": 6.545084971874738e-05, | |
| "loss": 0.2276, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.373015873015873, | |
| "grad_norm": 0.3670620322227478, | |
| "learning_rate": 6.378186779084995e-05, | |
| "loss": 0.2313, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.38095238095238093, | |
| "grad_norm": 0.29657959938049316, | |
| "learning_rate": 6.209609477998338e-05, | |
| "loss": 0.2078, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.3888888888888889, | |
| "grad_norm": 0.40728524327278137, | |
| "learning_rate": 6.0395584540887963e-05, | |
| "loss": 0.1976, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.3968253968253968, | |
| "grad_norm": 0.3712082803249359, | |
| "learning_rate": 5.868240888334653e-05, | |
| "loss": 0.2026, | |
| "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": 3.894908238692352e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |