| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.04067314047485891, | |
| "eval_steps": 13, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0008134628094971783, | |
| "grad_norm": 0.9108312726020813, | |
| "learning_rate": 1e-05, | |
| "loss": 1.4242, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0008134628094971783, | |
| "eval_loss": 1.6578611135482788, | |
| "eval_runtime": 195.6942, | |
| "eval_samples_per_second": 10.583, | |
| "eval_steps_per_second": 5.294, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0016269256189943566, | |
| "grad_norm": 0.8185848593711853, | |
| "learning_rate": 2e-05, | |
| "loss": 1.4684, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.0024403884284915347, | |
| "grad_norm": 0.8753907680511475, | |
| "learning_rate": 3e-05, | |
| "loss": 1.4355, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0032538512379887133, | |
| "grad_norm": 0.8084831237792969, | |
| "learning_rate": 4e-05, | |
| "loss": 1.4623, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.004067314047485892, | |
| "grad_norm": 0.776229202747345, | |
| "learning_rate": 5e-05, | |
| "loss": 1.4744, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.0048807768569830694, | |
| "grad_norm": 0.7320299744606018, | |
| "learning_rate": 6e-05, | |
| "loss": 1.5194, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.005694239666480248, | |
| "grad_norm": 0.8332337141036987, | |
| "learning_rate": 7e-05, | |
| "loss": 1.4006, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.0065077024759774265, | |
| "grad_norm": 0.9180299639701843, | |
| "learning_rate": 8e-05, | |
| "loss": 1.4191, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.007321165285474605, | |
| "grad_norm": 0.8599632382392883, | |
| "learning_rate": 9e-05, | |
| "loss": 1.3284, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.008134628094971784, | |
| "grad_norm": 1.0169434547424316, | |
| "learning_rate": 0.0001, | |
| "loss": 1.3026, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.008948090904468962, | |
| "grad_norm": 1.1758923530578613, | |
| "learning_rate": 9.98458666866564e-05, | |
| "loss": 1.3743, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.009761553713966139, | |
| "grad_norm": 1.301285982131958, | |
| "learning_rate": 9.938441702975689e-05, | |
| "loss": 1.3781, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.010575016523463317, | |
| "grad_norm": 1.1804258823394775, | |
| "learning_rate": 9.861849601988383e-05, | |
| "loss": 1.4496, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.010575016523463317, | |
| "eval_loss": 1.3726698160171509, | |
| "eval_runtime": 36.4237, | |
| "eval_samples_per_second": 56.859, | |
| "eval_steps_per_second": 28.443, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.011388479332960496, | |
| "grad_norm": 1.139792799949646, | |
| "learning_rate": 9.755282581475769e-05, | |
| "loss": 1.2662, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.012201942142457674, | |
| "grad_norm": 1.2657827138900757, | |
| "learning_rate": 9.619397662556435e-05, | |
| "loss": 1.2368, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.013015404951954853, | |
| "grad_norm": 0.998916745185852, | |
| "learning_rate": 9.45503262094184e-05, | |
| "loss": 1.1436, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.013828867761452032, | |
| "grad_norm": 1.0009139776229858, | |
| "learning_rate": 9.263200821770461e-05, | |
| "loss": 1.239, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.01464233057094921, | |
| "grad_norm": 1.1696265935897827, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 1.2521, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.015455793380446387, | |
| "grad_norm": 0.9146084189414978, | |
| "learning_rate": 8.802029828000156e-05, | |
| "loss": 1.1897, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.016269256189943567, | |
| "grad_norm": 0.9100729823112488, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 1.084, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.017082718999440746, | |
| "grad_norm": 0.8398633599281311, | |
| "learning_rate": 8.247240241650918e-05, | |
| "loss": 1.2046, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.017896181808937924, | |
| "grad_norm": 0.8264802694320679, | |
| "learning_rate": 7.938926261462366e-05, | |
| "loss": 1.2017, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.0187096446184351, | |
| "grad_norm": 0.8314357399940491, | |
| "learning_rate": 7.612492823579745e-05, | |
| "loss": 1.2994, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.019523107427932278, | |
| "grad_norm": 0.9849143028259277, | |
| "learning_rate": 7.269952498697734e-05, | |
| "loss": 1.363, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.020336570237429456, | |
| "grad_norm": 1.168421983718872, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 1.2528, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.021150033046926635, | |
| "grad_norm": 1.078614354133606, | |
| "learning_rate": 6.545084971874738e-05, | |
| "loss": 1.3879, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.021150033046926635, | |
| "eval_loss": 1.2557233572006226, | |
| "eval_runtime": 36.435, | |
| "eval_samples_per_second": 56.841, | |
| "eval_steps_per_second": 28.434, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.021963495856423813, | |
| "grad_norm": 0.7654521465301514, | |
| "learning_rate": 6.167226819279528e-05, | |
| "loss": 1.1113, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.022776958665920992, | |
| "grad_norm": 0.7602103352546692, | |
| "learning_rate": 5.782172325201155e-05, | |
| "loss": 1.2211, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.02359042147541817, | |
| "grad_norm": 0.8696973323822021, | |
| "learning_rate": 5.392295478639225e-05, | |
| "loss": 1.1937, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.02440388428491535, | |
| "grad_norm": 0.6972202062606812, | |
| "learning_rate": 5e-05, | |
| "loss": 1.1481, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.025217347094412527, | |
| "grad_norm": 0.8415080308914185, | |
| "learning_rate": 4.607704521360776e-05, | |
| "loss": 1.3055, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.026030809903909706, | |
| "grad_norm": 0.6529223322868347, | |
| "learning_rate": 4.2178276747988446e-05, | |
| "loss": 1.2149, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.026844272713406885, | |
| "grad_norm": 0.8038480281829834, | |
| "learning_rate": 3.832773180720475e-05, | |
| "loss": 1.1512, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.027657735522904063, | |
| "grad_norm": 0.7951949238777161, | |
| "learning_rate": 3.4549150281252636e-05, | |
| "loss": 1.2931, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.02847119833240124, | |
| "grad_norm": 0.6698573231697083, | |
| "learning_rate": 3.086582838174551e-05, | |
| "loss": 1.1721, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.02928466114189842, | |
| "grad_norm": 0.7140055298805237, | |
| "learning_rate": 2.7300475013022663e-05, | |
| "loss": 1.1752, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.0300981239513956, | |
| "grad_norm": 0.6723005175590515, | |
| "learning_rate": 2.3875071764202563e-05, | |
| "loss": 1.2337, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.030911586760892774, | |
| "grad_norm": 0.6740083694458008, | |
| "learning_rate": 2.061073738537635e-05, | |
| "loss": 1.2048, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.031725049570389956, | |
| "grad_norm": 0.7262203097343445, | |
| "learning_rate": 1.7527597583490822e-05, | |
| "loss": 1.3243, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.031725049570389956, | |
| "eval_loss": 1.2246425151824951, | |
| "eval_runtime": 36.4241, | |
| "eval_samples_per_second": 56.858, | |
| "eval_steps_per_second": 28.443, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.032538512379887134, | |
| "grad_norm": 0.7718028426170349, | |
| "learning_rate": 1.4644660940672627e-05, | |
| "loss": 1.2117, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.03335197518938431, | |
| "grad_norm": 0.7411465048789978, | |
| "learning_rate": 1.1979701719998453e-05, | |
| "loss": 1.1954, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.03416543799888149, | |
| "grad_norm": 0.7557942867279053, | |
| "learning_rate": 9.549150281252633e-06, | |
| "loss": 1.1628, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.03497890080837867, | |
| "grad_norm": 0.9376694560050964, | |
| "learning_rate": 7.367991782295391e-06, | |
| "loss": 1.2205, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.03579236361787585, | |
| "grad_norm": 0.7513824105262756, | |
| "learning_rate": 5.449673790581611e-06, | |
| "loss": 1.2009, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.03660582642737303, | |
| "grad_norm": 0.7382362484931946, | |
| "learning_rate": 3.8060233744356633e-06, | |
| "loss": 1.2458, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.0374192892368702, | |
| "grad_norm": 0.7333554625511169, | |
| "learning_rate": 2.4471741852423237e-06, | |
| "loss": 1.1459, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.03823275204636738, | |
| "grad_norm": 0.6778184175491333, | |
| "learning_rate": 1.3815039801161721e-06, | |
| "loss": 1.1253, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.039046214855864556, | |
| "grad_norm": 0.7480204105377197, | |
| "learning_rate": 6.15582970243117e-07, | |
| "loss": 1.2115, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.039859677665361734, | |
| "grad_norm": 0.7647523283958435, | |
| "learning_rate": 1.5413331334360182e-07, | |
| "loss": 1.2324, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.04067314047485891, | |
| "grad_norm": 0.8021819591522217, | |
| "learning_rate": 0.0, | |
| "loss": 1.2514, | |
| "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": 1.53307661205504e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |