| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.13506668917778153, | |
| "eval_steps": 13, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0027013337835556307, | |
| "grad_norm": 0.18103669583797455, | |
| "learning_rate": 1e-05, | |
| "loss": 1.4768, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0027013337835556307, | |
| "eval_loss": 1.515325665473938, | |
| "eval_runtime": 102.7627, | |
| "eval_samples_per_second": 6.072, | |
| "eval_steps_per_second": 3.036, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0054026675671112615, | |
| "grad_norm": 0.19400662183761597, | |
| "learning_rate": 2e-05, | |
| "loss": 1.4582, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.008104001350666891, | |
| "grad_norm": 0.14791804552078247, | |
| "learning_rate": 3e-05, | |
| "loss": 1.2738, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.010805335134222523, | |
| "grad_norm": 0.14204905927181244, | |
| "learning_rate": 4e-05, | |
| "loss": 1.3744, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.013506668917778153, | |
| "grad_norm": 0.13965387642383575, | |
| "learning_rate": 5e-05, | |
| "loss": 1.168, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.016208002701333782, | |
| "grad_norm": 0.1606331765651703, | |
| "learning_rate": 6e-05, | |
| "loss": 1.2615, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.018909336484889416, | |
| "grad_norm": 0.16424395143985748, | |
| "learning_rate": 7e-05, | |
| "loss": 1.2599, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.021610670268445046, | |
| "grad_norm": 0.22166331112384796, | |
| "learning_rate": 8e-05, | |
| "loss": 1.4082, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.024312004052000676, | |
| "grad_norm": 0.19696258008480072, | |
| "learning_rate": 9e-05, | |
| "loss": 1.3881, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.027013337835556307, | |
| "grad_norm": 0.23053531348705292, | |
| "learning_rate": 0.0001, | |
| "loss": 1.3014, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.029714671619111937, | |
| "grad_norm": 0.22696362435817719, | |
| "learning_rate": 9.98458666866564e-05, | |
| "loss": 1.3396, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.032416005402667564, | |
| "grad_norm": 0.19953660666942596, | |
| "learning_rate": 9.938441702975689e-05, | |
| "loss": 1.164, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.0351173391862232, | |
| "grad_norm": 0.21297410130500793, | |
| "learning_rate": 9.861849601988383e-05, | |
| "loss": 1.3591, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.0351173391862232, | |
| "eval_loss": 1.3835619688034058, | |
| "eval_runtime": 21.4452, | |
| "eval_samples_per_second": 29.097, | |
| "eval_steps_per_second": 14.549, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.03781867296977883, | |
| "grad_norm": 0.21262897551059723, | |
| "learning_rate": 9.755282581475769e-05, | |
| "loss": 1.1952, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.04052000675333446, | |
| "grad_norm": 0.2228931188583374, | |
| "learning_rate": 9.619397662556435e-05, | |
| "loss": 1.2496, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.04322134053689009, | |
| "grad_norm": 0.1923513263463974, | |
| "learning_rate": 9.45503262094184e-05, | |
| "loss": 1.3136, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.04592267432044572, | |
| "grad_norm": 0.19064629077911377, | |
| "learning_rate": 9.263200821770461e-05, | |
| "loss": 1.1784, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.04862400810400135, | |
| "grad_norm": 0.1903553009033203, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 1.3337, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.05132534188755698, | |
| "grad_norm": 0.18997488915920258, | |
| "learning_rate": 8.802029828000156e-05, | |
| "loss": 1.1441, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.05402667567111261, | |
| "grad_norm": 0.2513015568256378, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 1.1431, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.05672800945466824, | |
| "grad_norm": 0.20887787640094757, | |
| "learning_rate": 8.247240241650918e-05, | |
| "loss": 1.1726, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.059429343238223874, | |
| "grad_norm": 0.22922278940677643, | |
| "learning_rate": 7.938926261462366e-05, | |
| "loss": 1.2364, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.062130677021779504, | |
| "grad_norm": 0.2258031815290451, | |
| "learning_rate": 7.612492823579745e-05, | |
| "loss": 1.0899, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.06483201080533513, | |
| "grad_norm": 0.237678661942482, | |
| "learning_rate": 7.269952498697734e-05, | |
| "loss": 1.091, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.06753334458889076, | |
| "grad_norm": 0.22877460718154907, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 1.1396, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.0702346783724464, | |
| "grad_norm": 0.2620464563369751, | |
| "learning_rate": 6.545084971874738e-05, | |
| "loss": 1.1179, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.0702346783724464, | |
| "eval_loss": 1.2547305822372437, | |
| "eval_runtime": 21.5256, | |
| "eval_samples_per_second": 28.989, | |
| "eval_steps_per_second": 14.494, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.07293601215600203, | |
| "grad_norm": 0.2152528017759323, | |
| "learning_rate": 6.167226819279528e-05, | |
| "loss": 1.2901, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.07563734593955766, | |
| "grad_norm": 0.22586144506931305, | |
| "learning_rate": 5.782172325201155e-05, | |
| "loss": 1.1882, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.07833867972311329, | |
| "grad_norm": 0.22725199162960052, | |
| "learning_rate": 5.392295478639225e-05, | |
| "loss": 1.1866, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.08104001350666892, | |
| "grad_norm": 0.20386184751987457, | |
| "learning_rate": 5e-05, | |
| "loss": 1.1657, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.08374134729022455, | |
| "grad_norm": 0.2280956506729126, | |
| "learning_rate": 4.607704521360776e-05, | |
| "loss": 1.2514, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.08644268107378018, | |
| "grad_norm": 0.210204616189003, | |
| "learning_rate": 4.2178276747988446e-05, | |
| "loss": 1.0071, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.08914401485733581, | |
| "grad_norm": 0.21837709844112396, | |
| "learning_rate": 3.832773180720475e-05, | |
| "loss": 1.147, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.09184534864089144, | |
| "grad_norm": 0.23454667627811432, | |
| "learning_rate": 3.4549150281252636e-05, | |
| "loss": 1.0993, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.09454668242444707, | |
| "grad_norm": 0.22570019960403442, | |
| "learning_rate": 3.086582838174551e-05, | |
| "loss": 1.1352, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.0972480162080027, | |
| "grad_norm": 0.20457398891448975, | |
| "learning_rate": 2.7300475013022663e-05, | |
| "loss": 1.1004, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.09994934999155833, | |
| "grad_norm": 0.20898611843585968, | |
| "learning_rate": 2.3875071764202563e-05, | |
| "loss": 1.0908, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.10265068377511397, | |
| "grad_norm": 0.21299485862255096, | |
| "learning_rate": 2.061073738537635e-05, | |
| "loss": 1.1544, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.10535201755866959, | |
| "grad_norm": 0.19731168448925018, | |
| "learning_rate": 1.7527597583490822e-05, | |
| "loss": 1.1738, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.10535201755866959, | |
| "eval_loss": 1.2369698286056519, | |
| "eval_runtime": 21.5205, | |
| "eval_samples_per_second": 28.996, | |
| "eval_steps_per_second": 14.498, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.10805335134222523, | |
| "grad_norm": 0.22584712505340576, | |
| "learning_rate": 1.4644660940672627e-05, | |
| "loss": 1.18, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.11075468512578085, | |
| "grad_norm": 0.22953017055988312, | |
| "learning_rate": 1.1979701719998453e-05, | |
| "loss": 1.2876, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.11345601890933649, | |
| "grad_norm": 0.22218628227710724, | |
| "learning_rate": 9.549150281252633e-06, | |
| "loss": 1.2246, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.11615735269289211, | |
| "grad_norm": 0.228598952293396, | |
| "learning_rate": 7.367991782295391e-06, | |
| "loss": 1.2262, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.11885868647644775, | |
| "grad_norm": 0.2286594957113266, | |
| "learning_rate": 5.449673790581611e-06, | |
| "loss": 1.1715, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.12156002026000337, | |
| "grad_norm": 0.2573833465576172, | |
| "learning_rate": 3.8060233744356633e-06, | |
| "loss": 1.2798, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.12426135404355901, | |
| "grad_norm": 0.19452853500843048, | |
| "learning_rate": 2.4471741852423237e-06, | |
| "loss": 1.3053, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.12696268782711465, | |
| "grad_norm": 0.18400336802005768, | |
| "learning_rate": 1.3815039801161721e-06, | |
| "loss": 1.215, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.12966402161067025, | |
| "grad_norm": 0.20695160329341888, | |
| "learning_rate": 6.15582970243117e-07, | |
| "loss": 1.1684, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.1323653553942259, | |
| "grad_norm": 0.2696451246738434, | |
| "learning_rate": 1.5413331334360182e-07, | |
| "loss": 1.2025, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.13506668917778153, | |
| "grad_norm": 0.18349821865558624, | |
| "learning_rate": 0.0, | |
| "loss": 1.1954, | |
| "step": 50 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 50, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 13, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 5.73327259163689e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |