| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.0327653997378768, | |
| "eval_steps": 13, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.000655307994757536, | |
| "grad_norm": 0.016265055164694786, | |
| "learning_rate": 1e-05, | |
| "loss": 11.9312, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.000655307994757536, | |
| "eval_loss": 11.931286811828613, | |
| "eval_runtime": 28.0473, | |
| "eval_samples_per_second": 91.667, | |
| "eval_steps_per_second": 45.851, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.001310615989515072, | |
| "grad_norm": 0.014278396964073181, | |
| "learning_rate": 2e-05, | |
| "loss": 11.9309, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.001965923984272608, | |
| "grad_norm": 0.011991339735686779, | |
| "learning_rate": 3e-05, | |
| "loss": 11.9327, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.002621231979030144, | |
| "grad_norm": 0.01234712265431881, | |
| "learning_rate": 4e-05, | |
| "loss": 11.9315, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.00327653997378768, | |
| "grad_norm": 0.017012331634759903, | |
| "learning_rate": 5e-05, | |
| "loss": 11.9296, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.003931847968545216, | |
| "grad_norm": 0.014782086946070194, | |
| "learning_rate": 6e-05, | |
| "loss": 11.9329, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.0045871559633027525, | |
| "grad_norm": 0.020223338156938553, | |
| "learning_rate": 7e-05, | |
| "loss": 11.9317, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.005242463958060288, | |
| "grad_norm": 0.013544556684792042, | |
| "learning_rate": 8e-05, | |
| "loss": 11.9312, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.005897771952817824, | |
| "grad_norm": 0.012503627687692642, | |
| "learning_rate": 9e-05, | |
| "loss": 11.9315, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.00655307994757536, | |
| "grad_norm": 0.012558264657855034, | |
| "learning_rate": 0.0001, | |
| "loss": 11.9305, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.007208387942332897, | |
| "grad_norm": 0.015282824635505676, | |
| "learning_rate": 9.98458666866564e-05, | |
| "loss": 11.9317, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.007863695937090432, | |
| "grad_norm": 0.01448417454957962, | |
| "learning_rate": 9.938441702975689e-05, | |
| "loss": 11.9314, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.00851900393184797, | |
| "grad_norm": 0.012158408761024475, | |
| "learning_rate": 9.861849601988383e-05, | |
| "loss": 11.9323, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.00851900393184797, | |
| "eval_loss": 11.93112564086914, | |
| "eval_runtime": 12.5386, | |
| "eval_samples_per_second": 205.047, | |
| "eval_steps_per_second": 102.563, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.009174311926605505, | |
| "grad_norm": 0.011835216544568539, | |
| "learning_rate": 9.755282581475769e-05, | |
| "loss": 11.9299, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.009829619921363041, | |
| "grad_norm": 0.01589955948293209, | |
| "learning_rate": 9.619397662556435e-05, | |
| "loss": 11.9325, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.010484927916120577, | |
| "grad_norm": 0.010300605557858944, | |
| "learning_rate": 9.45503262094184e-05, | |
| "loss": 11.93, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.011140235910878113, | |
| "grad_norm": 0.013794222846627235, | |
| "learning_rate": 9.263200821770461e-05, | |
| "loss": 11.9315, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.011795543905635648, | |
| "grad_norm": 0.014298832975327969, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 11.9313, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.012450851900393184, | |
| "grad_norm": 0.017451252788305283, | |
| "learning_rate": 8.802029828000156e-05, | |
| "loss": 11.9296, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.01310615989515072, | |
| "grad_norm": 0.017337026074528694, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 11.9303, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.013761467889908258, | |
| "grad_norm": 0.016705993562936783, | |
| "learning_rate": 8.247240241650918e-05, | |
| "loss": 11.9319, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.014416775884665793, | |
| "grad_norm": 0.014467249624431133, | |
| "learning_rate": 7.938926261462366e-05, | |
| "loss": 11.9311, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.01507208387942333, | |
| "grad_norm": 0.017065448686480522, | |
| "learning_rate": 7.612492823579745e-05, | |
| "loss": 11.9321, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.015727391874180863, | |
| "grad_norm": 0.014603197574615479, | |
| "learning_rate": 7.269952498697734e-05, | |
| "loss": 11.9317, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.0163826998689384, | |
| "grad_norm": 0.020358189940452576, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 11.9313, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.01703800786369594, | |
| "grad_norm": 0.012381589971482754, | |
| "learning_rate": 6.545084971874738e-05, | |
| "loss": 11.9309, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.01703800786369594, | |
| "eval_loss": 11.930882453918457, | |
| "eval_runtime": 12.5197, | |
| "eval_samples_per_second": 205.357, | |
| "eval_steps_per_second": 102.718, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.017693315858453473, | |
| "grad_norm": 0.014943009242415428, | |
| "learning_rate": 6.167226819279528e-05, | |
| "loss": 11.9304, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.01834862385321101, | |
| "grad_norm": 0.02193116582930088, | |
| "learning_rate": 5.782172325201155e-05, | |
| "loss": 11.9322, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.019003931847968544, | |
| "grad_norm": 0.012562782503664494, | |
| "learning_rate": 5.392295478639225e-05, | |
| "loss": 11.9306, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.019659239842726082, | |
| "grad_norm": 0.013389959931373596, | |
| "learning_rate": 5e-05, | |
| "loss": 11.9316, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.020314547837483616, | |
| "grad_norm": 0.01724652200937271, | |
| "learning_rate": 4.607704521360776e-05, | |
| "loss": 11.9305, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.020969855832241154, | |
| "grad_norm": 0.018411165103316307, | |
| "learning_rate": 4.2178276747988446e-05, | |
| "loss": 11.9313, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.02162516382699869, | |
| "grad_norm": 0.013785377144813538, | |
| "learning_rate": 3.832773180720475e-05, | |
| "loss": 11.9316, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.022280471821756225, | |
| "grad_norm": 0.018790163099765778, | |
| "learning_rate": 3.4549150281252636e-05, | |
| "loss": 11.9323, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.022935779816513763, | |
| "grad_norm": 0.015211019665002823, | |
| "learning_rate": 3.086582838174551e-05, | |
| "loss": 11.9303, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.023591087811271297, | |
| "grad_norm": 0.01657300628721714, | |
| "learning_rate": 2.7300475013022663e-05, | |
| "loss": 11.9292, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.024246395806028834, | |
| "grad_norm": 0.01142768282443285, | |
| "learning_rate": 2.3875071764202563e-05, | |
| "loss": 11.9314, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.02490170380078637, | |
| "grad_norm": 0.021842923015356064, | |
| "learning_rate": 2.061073738537635e-05, | |
| "loss": 11.9316, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.025557011795543906, | |
| "grad_norm": 0.019510934129357338, | |
| "learning_rate": 1.7527597583490822e-05, | |
| "loss": 11.9325, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.025557011795543906, | |
| "eval_loss": 11.930744171142578, | |
| "eval_runtime": 12.5232, | |
| "eval_samples_per_second": 205.298, | |
| "eval_steps_per_second": 102.689, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.02621231979030144, | |
| "grad_norm": 0.013880622573196888, | |
| "learning_rate": 1.4644660940672627e-05, | |
| "loss": 11.9314, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.026867627785058978, | |
| "grad_norm": 0.019777296110987663, | |
| "learning_rate": 1.1979701719998453e-05, | |
| "loss": 11.9305, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.027522935779816515, | |
| "grad_norm": 0.013822672888636589, | |
| "learning_rate": 9.549150281252633e-06, | |
| "loss": 11.9314, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.02817824377457405, | |
| "grad_norm": 0.016023075208067894, | |
| "learning_rate": 7.367991782295391e-06, | |
| "loss": 11.9313, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.028833551769331587, | |
| "grad_norm": 0.01633632369339466, | |
| "learning_rate": 5.449673790581611e-06, | |
| "loss": 11.931, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.02948885976408912, | |
| "grad_norm": 0.014971333555877209, | |
| "learning_rate": 3.8060233744356633e-06, | |
| "loss": 11.9318, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.03014416775884666, | |
| "grad_norm": 0.01602235995233059, | |
| "learning_rate": 2.4471741852423237e-06, | |
| "loss": 11.9318, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.030799475753604193, | |
| "grad_norm": 0.013982513919472694, | |
| "learning_rate": 1.3815039801161721e-06, | |
| "loss": 11.9317, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.03145478374836173, | |
| "grad_norm": 0.01966976933181286, | |
| "learning_rate": 6.15582970243117e-07, | |
| "loss": 11.9299, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.03211009174311927, | |
| "grad_norm": 0.018536817282438278, | |
| "learning_rate": 1.5413331334360182e-07, | |
| "loss": 11.9314, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.0327653997378768, | |
| "grad_norm": 0.01766209863126278, | |
| "learning_rate": 0.0, | |
| "loss": 11.9292, | |
| "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": 299787878400.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |