| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.03251239535072747, | |
| "eval_steps": 13, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0006502479070145493, | |
| "grad_norm": 0.16097666323184967, | |
| "learning_rate": 1e-05, | |
| "loss": 1.2708, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0006502479070145493, | |
| "eval_loss": 1.3661339282989502, | |
| "eval_runtime": 227.5771, | |
| "eval_samples_per_second": 11.385, | |
| "eval_steps_per_second": 5.695, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0013004958140290985, | |
| "grad_norm": 0.19011889398097992, | |
| "learning_rate": 2e-05, | |
| "loss": 1.355, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.001950743721043648, | |
| "grad_norm": 0.16794496774673462, | |
| "learning_rate": 3e-05, | |
| "loss": 1.3159, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.002600991628058197, | |
| "grad_norm": 0.15826600790023804, | |
| "learning_rate": 4e-05, | |
| "loss": 1.327, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.0032512395350727465, | |
| "grad_norm": 0.1772988885641098, | |
| "learning_rate": 5e-05, | |
| "loss": 1.2694, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.003901487442087296, | |
| "grad_norm": 0.16691641509532928, | |
| "learning_rate": 6e-05, | |
| "loss": 1.2382, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.004551735349101845, | |
| "grad_norm": 0.17923057079315186, | |
| "learning_rate": 7e-05, | |
| "loss": 1.2678, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.005201983256116394, | |
| "grad_norm": 0.19407686591148376, | |
| "learning_rate": 8e-05, | |
| "loss": 1.2386, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.005852231163130944, | |
| "grad_norm": 0.21998471021652222, | |
| "learning_rate": 9e-05, | |
| "loss": 1.3607, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.006502479070145493, | |
| "grad_norm": 0.2332504242658615, | |
| "learning_rate": 0.0001, | |
| "loss": 1.3526, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.007152726977160042, | |
| "grad_norm": 0.20302407443523407, | |
| "learning_rate": 9.98458666866564e-05, | |
| "loss": 1.2525, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.007802974884174592, | |
| "grad_norm": 0.20793932676315308, | |
| "learning_rate": 9.938441702975689e-05, | |
| "loss": 1.3305, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.00845322279118914, | |
| "grad_norm": 0.24406825006008148, | |
| "learning_rate": 9.861849601988383e-05, | |
| "loss": 1.3575, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.00845322279118914, | |
| "eval_loss": 1.287429928779602, | |
| "eval_runtime": 74.3569, | |
| "eval_samples_per_second": 34.845, | |
| "eval_steps_per_second": 17.429, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.00910347069820369, | |
| "grad_norm": 0.24397294223308563, | |
| "learning_rate": 9.755282581475769e-05, | |
| "loss": 1.2718, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.00975371860521824, | |
| "grad_norm": 0.2642868459224701, | |
| "learning_rate": 9.619397662556435e-05, | |
| "loss": 1.2672, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.010403966512232788, | |
| "grad_norm": 0.29374638199806213, | |
| "learning_rate": 9.45503262094184e-05, | |
| "loss": 1.2364, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.011054214419247338, | |
| "grad_norm": 0.2562926709651947, | |
| "learning_rate": 9.263200821770461e-05, | |
| "loss": 1.2337, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.011704462326261888, | |
| "grad_norm": 0.24123510718345642, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 1.2308, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.012354710233276436, | |
| "grad_norm": 0.298176109790802, | |
| "learning_rate": 8.802029828000156e-05, | |
| "loss": 1.269, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.013004958140290986, | |
| "grad_norm": 0.29537469148635864, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 1.2055, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.013655206047305536, | |
| "grad_norm": 0.2935850918292999, | |
| "learning_rate": 8.247240241650918e-05, | |
| "loss": 1.2187, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.014305453954320084, | |
| "grad_norm": 0.27675747871398926, | |
| "learning_rate": 7.938926261462366e-05, | |
| "loss": 1.2315, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.014955701861334634, | |
| "grad_norm": 0.2816616892814636, | |
| "learning_rate": 7.612492823579745e-05, | |
| "loss": 1.1898, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.015605949768349184, | |
| "grad_norm": 0.25230491161346436, | |
| "learning_rate": 7.269952498697734e-05, | |
| "loss": 1.2143, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.016256197675363734, | |
| "grad_norm": 0.31681883335113525, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 1.2153, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.01690644558237828, | |
| "grad_norm": 0.27878355979919434, | |
| "learning_rate": 6.545084971874738e-05, | |
| "loss": 1.1783, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.01690644558237828, | |
| "eval_loss": 1.2144542932510376, | |
| "eval_runtime": 74.3264, | |
| "eval_samples_per_second": 34.86, | |
| "eval_steps_per_second": 17.437, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.01755669348939283, | |
| "grad_norm": 0.29930874705314636, | |
| "learning_rate": 6.167226819279528e-05, | |
| "loss": 1.2125, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.01820694139640738, | |
| "grad_norm": 0.27576595544815063, | |
| "learning_rate": 5.782172325201155e-05, | |
| "loss": 1.1358, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.01885718930342193, | |
| "grad_norm": 0.34936976432800293, | |
| "learning_rate": 5.392295478639225e-05, | |
| "loss": 1.2408, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.01950743721043648, | |
| "grad_norm": 0.32189854979515076, | |
| "learning_rate": 5e-05, | |
| "loss": 1.2407, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.02015768511745103, | |
| "grad_norm": 0.31609469652175903, | |
| "learning_rate": 4.607704521360776e-05, | |
| "loss": 1.213, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.020807933024465576, | |
| "grad_norm": 0.36792436242103577, | |
| "learning_rate": 4.2178276747988446e-05, | |
| "loss": 1.24, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.021458180931480126, | |
| "grad_norm": 0.293441504240036, | |
| "learning_rate": 3.832773180720475e-05, | |
| "loss": 1.1205, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.022108428838494676, | |
| "grad_norm": 0.357262521982193, | |
| "learning_rate": 3.4549150281252636e-05, | |
| "loss": 1.2352, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.022758676745509226, | |
| "grad_norm": 0.35729458928108215, | |
| "learning_rate": 3.086582838174551e-05, | |
| "loss": 1.0767, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.023408924652523776, | |
| "grad_norm": 0.3217020034790039, | |
| "learning_rate": 2.7300475013022663e-05, | |
| "loss": 1.2367, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.024059172559538326, | |
| "grad_norm": 0.3059484362602234, | |
| "learning_rate": 2.3875071764202563e-05, | |
| "loss": 1.1982, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.024709420466552872, | |
| "grad_norm": 0.3412172198295593, | |
| "learning_rate": 2.061073738537635e-05, | |
| "loss": 1.2419, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.025359668373567422, | |
| "grad_norm": 0.3185587227344513, | |
| "learning_rate": 1.7527597583490822e-05, | |
| "loss": 1.2128, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.025359668373567422, | |
| "eval_loss": 1.198966145515442, | |
| "eval_runtime": 74.3589, | |
| "eval_samples_per_second": 34.845, | |
| "eval_steps_per_second": 17.429, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.026009916280581972, | |
| "grad_norm": 0.3025207817554474, | |
| "learning_rate": 1.4644660940672627e-05, | |
| "loss": 1.2036, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.02666016418759652, | |
| "grad_norm": 0.3241982161998749, | |
| "learning_rate": 1.1979701719998453e-05, | |
| "loss": 1.2358, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.02731041209461107, | |
| "grad_norm": 0.37644854187965393, | |
| "learning_rate": 9.549150281252633e-06, | |
| "loss": 1.2374, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.027960660001625618, | |
| "grad_norm": 0.32864895462989807, | |
| "learning_rate": 7.367991782295391e-06, | |
| "loss": 1.1828, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.028610907908640168, | |
| "grad_norm": 0.3057841956615448, | |
| "learning_rate": 5.449673790581611e-06, | |
| "loss": 1.0924, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.029261155815654718, | |
| "grad_norm": 0.30859678983688354, | |
| "learning_rate": 3.8060233744356633e-06, | |
| "loss": 1.2563, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.029911403722669268, | |
| "grad_norm": 0.2911403775215149, | |
| "learning_rate": 2.4471741852423237e-06, | |
| "loss": 1.1663, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.030561651629683818, | |
| "grad_norm": 0.2932097017765045, | |
| "learning_rate": 1.3815039801161721e-06, | |
| "loss": 1.2638, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.031211899536698368, | |
| "grad_norm": 0.32970038056373596, | |
| "learning_rate": 6.15582970243117e-07, | |
| "loss": 1.2205, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.031862147443712914, | |
| "grad_norm": 0.3031351566314697, | |
| "learning_rate": 1.5413331334360182e-07, | |
| "loss": 1.1235, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.03251239535072747, | |
| "grad_norm": 0.3202054500579834, | |
| "learning_rate": 0.0, | |
| "loss": 1.2213, | |
| "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": 4.696404144213197e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |