| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.1343408900083963, | |
| "eval_steps": 13, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.002686817800167926, | |
| "grad_norm": 4.219735622406006, | |
| "learning_rate": 1e-05, | |
| "loss": 5.2831, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.002686817800167926, | |
| "eval_loss": 5.458459377288818, | |
| "eval_runtime": 100.9838, | |
| "eval_samples_per_second": 6.209, | |
| "eval_steps_per_second": 3.109, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.005373635600335852, | |
| "grad_norm": 4.188349723815918, | |
| "learning_rate": 2e-05, | |
| "loss": 5.2641, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.008060453400503778, | |
| "grad_norm": 3.9965884685516357, | |
| "learning_rate": 3e-05, | |
| "loss": 5.2305, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.010747271200671704, | |
| "grad_norm": 4.113009929656982, | |
| "learning_rate": 4e-05, | |
| "loss": 5.198, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.01343408900083963, | |
| "grad_norm": 3.996140480041504, | |
| "learning_rate": 5e-05, | |
| "loss": 5.0664, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.016120906801007556, | |
| "grad_norm": 4.285845756530762, | |
| "learning_rate": 6e-05, | |
| "loss": 5.1741, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.018807724601175482, | |
| "grad_norm": 4.329155445098877, | |
| "learning_rate": 7e-05, | |
| "loss": 4.9143, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.02149454240134341, | |
| "grad_norm": 4.412178039550781, | |
| "learning_rate": 8e-05, | |
| "loss": 4.6129, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.024181360201511334, | |
| "grad_norm": 4.658452033996582, | |
| "learning_rate": 9e-05, | |
| "loss": 4.5596, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.02686817800167926, | |
| "grad_norm": 4.991731643676758, | |
| "learning_rate": 0.0001, | |
| "loss": 4.1628, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.029554995801847186, | |
| "grad_norm": 5.284714221954346, | |
| "learning_rate": 9.98458666866564e-05, | |
| "loss": 4.0111, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.03224181360201511, | |
| "grad_norm": 6.26345157623291, | |
| "learning_rate": 9.938441702975689e-05, | |
| "loss": 3.9997, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.03492863140218304, | |
| "grad_norm": 6.786696910858154, | |
| "learning_rate": 9.861849601988383e-05, | |
| "loss": 3.7994, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.03492863140218304, | |
| "eval_loss": 3.7471344470977783, | |
| "eval_runtime": 16.0515, | |
| "eval_samples_per_second": 39.062, | |
| "eval_steps_per_second": 19.562, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.037615449202350965, | |
| "grad_norm": 6.086618423461914, | |
| "learning_rate": 9.755282581475769e-05, | |
| "loss": 3.6493, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.04030226700251889, | |
| "grad_norm": 7.451858043670654, | |
| "learning_rate": 9.619397662556435e-05, | |
| "loss": 3.6744, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.04298908480268682, | |
| "grad_norm": 6.979912757873535, | |
| "learning_rate": 9.45503262094184e-05, | |
| "loss": 3.4717, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.04567590260285474, | |
| "grad_norm": 7.578520774841309, | |
| "learning_rate": 9.263200821770461e-05, | |
| "loss": 3.2947, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.04836272040302267, | |
| "grad_norm": 8.897782325744629, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 3.2721, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.051049538203190595, | |
| "grad_norm": 7.789557456970215, | |
| "learning_rate": 8.802029828000156e-05, | |
| "loss": 3.0338, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.05373635600335852, | |
| "grad_norm": 8.348186492919922, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 2.9377, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.05642317380352645, | |
| "grad_norm": 9.296662330627441, | |
| "learning_rate": 8.247240241650918e-05, | |
| "loss": 2.9852, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.05910999160369437, | |
| "grad_norm": 8.833761215209961, | |
| "learning_rate": 7.938926261462366e-05, | |
| "loss": 2.8072, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.0617968094038623, | |
| "grad_norm": 8.135382652282715, | |
| "learning_rate": 7.612492823579745e-05, | |
| "loss": 2.5832, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.06448362720403022, | |
| "grad_norm": 7.976484298706055, | |
| "learning_rate": 7.269952498697734e-05, | |
| "loss": 2.44, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.06717044500419815, | |
| "grad_norm": 1.7394722700119019, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 2.4148, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.06985726280436608, | |
| "grad_norm": 1.333323359489441, | |
| "learning_rate": 6.545084971874738e-05, | |
| "loss": 2.3852, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.06985726280436608, | |
| "eval_loss": 2.4600071907043457, | |
| "eval_runtime": 16.0246, | |
| "eval_samples_per_second": 39.127, | |
| "eval_steps_per_second": 19.595, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.072544080604534, | |
| "grad_norm": 1.1186634302139282, | |
| "learning_rate": 6.167226819279528e-05, | |
| "loss": 2.3918, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.07523089840470193, | |
| "grad_norm": 1.239710807800293, | |
| "learning_rate": 5.782172325201155e-05, | |
| "loss": 2.3986, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.07791771620486986, | |
| "grad_norm": 1.718079686164856, | |
| "learning_rate": 5.392295478639225e-05, | |
| "loss": 2.3816, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.08060453400503778, | |
| "grad_norm": 1.1218688488006592, | |
| "learning_rate": 5e-05, | |
| "loss": 2.2879, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.08329135180520571, | |
| "grad_norm": 1.3196156024932861, | |
| "learning_rate": 4.607704521360776e-05, | |
| "loss": 2.3083, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.08597816960537363, | |
| "grad_norm": 1.2913845777511597, | |
| "learning_rate": 4.2178276747988446e-05, | |
| "loss": 2.4363, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.08866498740554156, | |
| "grad_norm": 1.186840534210205, | |
| "learning_rate": 3.832773180720475e-05, | |
| "loss": 2.2568, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.09135180520570949, | |
| "grad_norm": 1.1380304098129272, | |
| "learning_rate": 3.4549150281252636e-05, | |
| "loss": 2.2333, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.09403862300587741, | |
| "grad_norm": 1.1696350574493408, | |
| "learning_rate": 3.086582838174551e-05, | |
| "loss": 2.1891, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.09672544080604534, | |
| "grad_norm": 1.155229926109314, | |
| "learning_rate": 2.7300475013022663e-05, | |
| "loss": 2.2588, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.09941225860621326, | |
| "grad_norm": 1.1977050304412842, | |
| "learning_rate": 2.3875071764202563e-05, | |
| "loss": 2.3242, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.10209907640638119, | |
| "grad_norm": 1.2246061563491821, | |
| "learning_rate": 2.061073738537635e-05, | |
| "loss": 2.3783, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.10478589420654912, | |
| "grad_norm": 1.1910125017166138, | |
| "learning_rate": 1.7527597583490822e-05, | |
| "loss": 2.2939, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.10478589420654912, | |
| "eval_loss": 2.2964117527008057, | |
| "eval_runtime": 16.0505, | |
| "eval_samples_per_second": 39.064, | |
| "eval_steps_per_second": 19.563, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.10747271200671704, | |
| "grad_norm": 1.1191169023513794, | |
| "learning_rate": 1.4644660940672627e-05, | |
| "loss": 2.2467, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.11015952980688497, | |
| "grad_norm": 1.0125330686569214, | |
| "learning_rate": 1.1979701719998453e-05, | |
| "loss": 2.1799, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.1128463476070529, | |
| "grad_norm": 1.3123078346252441, | |
| "learning_rate": 9.549150281252633e-06, | |
| "loss": 2.2992, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.11553316540722082, | |
| "grad_norm": 1.011380672454834, | |
| "learning_rate": 7.367991782295391e-06, | |
| "loss": 2.2407, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.11821998320738875, | |
| "grad_norm": 1.1584933996200562, | |
| "learning_rate": 5.449673790581611e-06, | |
| "loss": 2.3031, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.12090680100755667, | |
| "grad_norm": 1.1705843210220337, | |
| "learning_rate": 3.8060233744356633e-06, | |
| "loss": 2.2667, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.1235936188077246, | |
| "grad_norm": 1.0741093158721924, | |
| "learning_rate": 2.4471741852423237e-06, | |
| "loss": 2.2942, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.12628043660789254, | |
| "grad_norm": 1.0633755922317505, | |
| "learning_rate": 1.3815039801161721e-06, | |
| "loss": 2.3064, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.12896725440806045, | |
| "grad_norm": 1.2350645065307617, | |
| "learning_rate": 6.15582970243117e-07, | |
| "loss": 2.2901, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.1316540722082284, | |
| "grad_norm": 1.0202724933624268, | |
| "learning_rate": 1.5413331334360182e-07, | |
| "loss": 2.187, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.1343408900083963, | |
| "grad_norm": 1.1233915090560913, | |
| "learning_rate": 0.0, | |
| "loss": 2.2938, | |
| "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": 3.8221828325376e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |