| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.18726591760299627, | |
| "eval_steps": 13, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.003745318352059925, | |
| "grad_norm": 1.4232933521270752, | |
| "learning_rate": 1e-05, | |
| "loss": 2.6144, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.003745318352059925, | |
| "eval_loss": 2.7996814250946045, | |
| "eval_runtime": 168.8282, | |
| "eval_samples_per_second": 2.665, | |
| "eval_steps_per_second": 1.333, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.00749063670411985, | |
| "grad_norm": 1.3964163064956665, | |
| "learning_rate": 2e-05, | |
| "loss": 2.6416, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.011235955056179775, | |
| "grad_norm": 1.4900565147399902, | |
| "learning_rate": 3e-05, | |
| "loss": 2.6927, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0149812734082397, | |
| "grad_norm": 1.2893002033233643, | |
| "learning_rate": 4e-05, | |
| "loss": 2.2962, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.018726591760299626, | |
| "grad_norm": 1.5299968719482422, | |
| "learning_rate": 5e-05, | |
| "loss": 2.997, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.02247191011235955, | |
| "grad_norm": 1.5312691926956177, | |
| "learning_rate": 6e-05, | |
| "loss": 2.8766, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.026217228464419477, | |
| "grad_norm": 1.6678813695907593, | |
| "learning_rate": 7e-05, | |
| "loss": 2.7168, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.0299625468164794, | |
| "grad_norm": 1.3376212120056152, | |
| "learning_rate": 8e-05, | |
| "loss": 2.2845, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.033707865168539325, | |
| "grad_norm": 1.4087949991226196, | |
| "learning_rate": 9e-05, | |
| "loss": 2.2124, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.03745318352059925, | |
| "grad_norm": 1.4170143604278564, | |
| "learning_rate": 0.0001, | |
| "loss": 2.3934, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.04119850187265917, | |
| "grad_norm": 1.392784833908081, | |
| "learning_rate": 9.98458666866564e-05, | |
| "loss": 2.1857, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.0449438202247191, | |
| "grad_norm": 1.4366754293441772, | |
| "learning_rate": 9.938441702975689e-05, | |
| "loss": 1.9748, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.04868913857677903, | |
| "grad_norm": 1.5409985780715942, | |
| "learning_rate": 9.861849601988383e-05, | |
| "loss": 1.9675, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.04868913857677903, | |
| "eval_loss": 1.975732684135437, | |
| "eval_runtime": 14.12, | |
| "eval_samples_per_second": 31.87, | |
| "eval_steps_per_second": 15.935, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.052434456928838954, | |
| "grad_norm": 1.596232533454895, | |
| "learning_rate": 9.755282581475769e-05, | |
| "loss": 1.8742, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.056179775280898875, | |
| "grad_norm": 1.6847773790359497, | |
| "learning_rate": 9.619397662556435e-05, | |
| "loss": 1.9345, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.0599250936329588, | |
| "grad_norm": 1.5066741704940796, | |
| "learning_rate": 9.45503262094184e-05, | |
| "loss": 1.7222, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.06367041198501873, | |
| "grad_norm": 1.4937152862548828, | |
| "learning_rate": 9.263200821770461e-05, | |
| "loss": 1.8033, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.06741573033707865, | |
| "grad_norm": 1.9379730224609375, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 1.8392, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.07116104868913857, | |
| "grad_norm": 1.7714548110961914, | |
| "learning_rate": 8.802029828000156e-05, | |
| "loss": 1.7254, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.0749063670411985, | |
| "grad_norm": 1.6995341777801514, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 1.5896, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.07865168539325842, | |
| "grad_norm": 1.6350839138031006, | |
| "learning_rate": 8.247240241650918e-05, | |
| "loss": 1.8164, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.08239700374531835, | |
| "grad_norm": 1.4452717304229736, | |
| "learning_rate": 7.938926261462366e-05, | |
| "loss": 1.7086, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.08614232209737828, | |
| "grad_norm": 1.4310646057128906, | |
| "learning_rate": 7.612492823579745e-05, | |
| "loss": 1.7038, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.0898876404494382, | |
| "grad_norm": 1.5116603374481201, | |
| "learning_rate": 7.269952498697734e-05, | |
| "loss": 1.3802, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.09363295880149813, | |
| "grad_norm": 1.7284044027328491, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 1.5872, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.09737827715355805, | |
| "grad_norm": 1.4467360973358154, | |
| "learning_rate": 6.545084971874738e-05, | |
| "loss": 1.464, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.09737827715355805, | |
| "eval_loss": 1.513379454612732, | |
| "eval_runtime": 14.1358, | |
| "eval_samples_per_second": 31.834, | |
| "eval_steps_per_second": 15.917, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.10112359550561797, | |
| "grad_norm": 1.6498875617980957, | |
| "learning_rate": 6.167226819279528e-05, | |
| "loss": 1.6355, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.10486891385767791, | |
| "grad_norm": 1.5848513841629028, | |
| "learning_rate": 5.782172325201155e-05, | |
| "loss": 1.4611, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.10861423220973783, | |
| "grad_norm": 1.6103304624557495, | |
| "learning_rate": 5.392295478639225e-05, | |
| "loss": 1.463, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.11235955056179775, | |
| "grad_norm": 1.4713010787963867, | |
| "learning_rate": 5e-05, | |
| "loss": 1.4531, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.11610486891385768, | |
| "grad_norm": 1.4418309926986694, | |
| "learning_rate": 4.607704521360776e-05, | |
| "loss": 1.4607, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.1198501872659176, | |
| "grad_norm": 1.3354123830795288, | |
| "learning_rate": 4.2178276747988446e-05, | |
| "loss": 1.4086, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.12359550561797752, | |
| "grad_norm": 1.5024315118789673, | |
| "learning_rate": 3.832773180720475e-05, | |
| "loss": 1.4571, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.12734082397003746, | |
| "grad_norm": 1.4268651008605957, | |
| "learning_rate": 3.4549150281252636e-05, | |
| "loss": 1.3551, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.13108614232209737, | |
| "grad_norm": 1.4591658115386963, | |
| "learning_rate": 3.086582838174551e-05, | |
| "loss": 1.3186, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.1348314606741573, | |
| "grad_norm": 1.5342679023742676, | |
| "learning_rate": 2.7300475013022663e-05, | |
| "loss": 1.1651, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.13857677902621723, | |
| "grad_norm": 1.763529896736145, | |
| "learning_rate": 2.3875071764202563e-05, | |
| "loss": 1.1908, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.14232209737827714, | |
| "grad_norm": 1.6287809610366821, | |
| "learning_rate": 2.061073738537635e-05, | |
| "loss": 1.4347, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.14606741573033707, | |
| "grad_norm": 1.562390923500061, | |
| "learning_rate": 1.7527597583490822e-05, | |
| "loss": 1.4056, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.14606741573033707, | |
| "eval_loss": 1.3499964475631714, | |
| "eval_runtime": 14.1251, | |
| "eval_samples_per_second": 31.858, | |
| "eval_steps_per_second": 15.929, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.149812734082397, | |
| "grad_norm": 1.6496071815490723, | |
| "learning_rate": 1.4644660940672627e-05, | |
| "loss": 1.1467, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.15355805243445692, | |
| "grad_norm": 1.5362358093261719, | |
| "learning_rate": 1.1979701719998453e-05, | |
| "loss": 1.3055, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.15730337078651685, | |
| "grad_norm": 1.656622052192688, | |
| "learning_rate": 9.549150281252633e-06, | |
| "loss": 1.2599, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.16104868913857678, | |
| "grad_norm": 1.7673784494400024, | |
| "learning_rate": 7.367991782295391e-06, | |
| "loss": 1.6338, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.1647940074906367, | |
| "grad_norm": 1.6738536357879639, | |
| "learning_rate": 5.449673790581611e-06, | |
| "loss": 1.3426, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.16853932584269662, | |
| "grad_norm": 1.5348960161209106, | |
| "learning_rate": 3.8060233744356633e-06, | |
| "loss": 1.3081, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.17228464419475656, | |
| "grad_norm": 1.4670664072036743, | |
| "learning_rate": 2.4471741852423237e-06, | |
| "loss": 1.2674, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.1760299625468165, | |
| "grad_norm": 1.5279347896575928, | |
| "learning_rate": 1.3815039801161721e-06, | |
| "loss": 1.3096, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.1797752808988764, | |
| "grad_norm": 1.749185562133789, | |
| "learning_rate": 6.15582970243117e-07, | |
| "loss": 1.3092, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.18352059925093633, | |
| "grad_norm": 1.499311089515686, | |
| "learning_rate": 1.5413331334360182e-07, | |
| "loss": 1.1741, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.18726591760299627, | |
| "grad_norm": 1.6856210231781006, | |
| "learning_rate": 0.0, | |
| "loss": 1.6819, | |
| "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": 4.284371676168192e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |