| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.11462960309499928, | |
| "eval_steps": 13, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0022925920618999856, | |
| "grad_norm": 11.200607299804688, | |
| "learning_rate": 1e-05, | |
| "loss": 11.4606, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0022925920618999856, | |
| "eval_loss": 11.853107452392578, | |
| "eval_runtime": 123.7045, | |
| "eval_samples_per_second": 5.942, | |
| "eval_steps_per_second": 2.975, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.004585184123799971, | |
| "grad_norm": 11.813929557800293, | |
| "learning_rate": 2e-05, | |
| "loss": 12.0419, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.006877776185699957, | |
| "grad_norm": 11.820816993713379, | |
| "learning_rate": 3e-05, | |
| "loss": 11.6056, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.009170368247599942, | |
| "grad_norm": 11.824806213378906, | |
| "learning_rate": 4e-05, | |
| "loss": 10.7151, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.011462960309499929, | |
| "grad_norm": 11.645264625549316, | |
| "learning_rate": 5e-05, | |
| "loss": 10.2699, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.013755552371399913, | |
| "grad_norm": 14.307526588439941, | |
| "learning_rate": 6e-05, | |
| "loss": 9.6184, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.0160481444332999, | |
| "grad_norm": 14.689462661743164, | |
| "learning_rate": 7e-05, | |
| "loss": 8.5351, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.018340736495199884, | |
| "grad_norm": 13.544893264770508, | |
| "learning_rate": 8e-05, | |
| "loss": 6.6334, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.020633328557099873, | |
| "grad_norm": 16.24001121520996, | |
| "learning_rate": 9e-05, | |
| "loss": 5.2598, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.022925920618999857, | |
| "grad_norm": 11.873700141906738, | |
| "learning_rate": 0.0001, | |
| "loss": 3.6346, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.025218512680899842, | |
| "grad_norm": 9.610759735107422, | |
| "learning_rate": 9.98458666866564e-05, | |
| "loss": 2.5759, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.027511104742799827, | |
| "grad_norm": 11.247403144836426, | |
| "learning_rate": 9.938441702975689e-05, | |
| "loss": 2.0404, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.029803696804699815, | |
| "grad_norm": 10.719569206237793, | |
| "learning_rate": 9.861849601988383e-05, | |
| "loss": 1.7831, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.029803696804699815, | |
| "eval_loss": 1.407051682472229, | |
| "eval_runtime": 107.3728, | |
| "eval_samples_per_second": 6.845, | |
| "eval_steps_per_second": 3.427, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.0320962888665998, | |
| "grad_norm": 13.326321601867676, | |
| "learning_rate": 9.755282581475769e-05, | |
| "loss": 1.4722, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.03438888092849979, | |
| "grad_norm": 6.82303524017334, | |
| "learning_rate": 9.619397662556435e-05, | |
| "loss": 1.0576, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.03668147299039977, | |
| "grad_norm": 7.098742961883545, | |
| "learning_rate": 9.45503262094184e-05, | |
| "loss": 0.9765, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.03897406505229976, | |
| "grad_norm": 8.935181617736816, | |
| "learning_rate": 9.263200821770461e-05, | |
| "loss": 1.1806, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.041266657114199745, | |
| "grad_norm": 8.649842262268066, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 0.8492, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.043559249176099726, | |
| "grad_norm": 8.415911674499512, | |
| "learning_rate": 8.802029828000156e-05, | |
| "loss": 0.8533, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.045851841237999715, | |
| "grad_norm": 8.740532875061035, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 1.0349, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.048144433299899696, | |
| "grad_norm": 4.766848564147949, | |
| "learning_rate": 8.247240241650918e-05, | |
| "loss": 0.6003, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.050437025361799684, | |
| "grad_norm": 3.6371521949768066, | |
| "learning_rate": 7.938926261462366e-05, | |
| "loss": 0.4889, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.05272961742369967, | |
| "grad_norm": 7.622241973876953, | |
| "learning_rate": 7.612492823579745e-05, | |
| "loss": 0.9877, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.05502220948559965, | |
| "grad_norm": 6.696074962615967, | |
| "learning_rate": 7.269952498697734e-05, | |
| "loss": 1.0169, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.05731480154749964, | |
| "grad_norm": 6.592733860015869, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 0.7066, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.05960739360939963, | |
| "grad_norm": 3.0361104011535645, | |
| "learning_rate": 6.545084971874738e-05, | |
| "loss": 0.5061, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.05960739360939963, | |
| "eval_loss": 0.5841758847236633, | |
| "eval_runtime": 107.4624, | |
| "eval_samples_per_second": 6.84, | |
| "eval_steps_per_second": 3.424, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.06189998567129961, | |
| "grad_norm": 2.6141648292541504, | |
| "learning_rate": 6.167226819279528e-05, | |
| "loss": 0.571, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.0641925777331996, | |
| "grad_norm": 3.1781935691833496, | |
| "learning_rate": 5.782172325201155e-05, | |
| "loss": 0.5977, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.06648516979509958, | |
| "grad_norm": 3.0517966747283936, | |
| "learning_rate": 5.392295478639225e-05, | |
| "loss": 0.4363, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.06877776185699958, | |
| "grad_norm": 4.312246322631836, | |
| "learning_rate": 5e-05, | |
| "loss": 0.5023, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.07107035391889956, | |
| "grad_norm": 3.547243118286133, | |
| "learning_rate": 4.607704521360776e-05, | |
| "loss": 0.5231, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.07336294598079954, | |
| "grad_norm": 3.709038257598877, | |
| "learning_rate": 4.2178276747988446e-05, | |
| "loss": 0.592, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.07565553804269953, | |
| "grad_norm": 3.735938310623169, | |
| "learning_rate": 3.832773180720475e-05, | |
| "loss": 0.6164, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.07794813010459951, | |
| "grad_norm": 2.0814919471740723, | |
| "learning_rate": 3.4549150281252636e-05, | |
| "loss": 0.4091, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.0802407221664995, | |
| "grad_norm": 3.5011098384857178, | |
| "learning_rate": 3.086582838174551e-05, | |
| "loss": 0.4432, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.08253331422839949, | |
| "grad_norm": 4.164271831512451, | |
| "learning_rate": 2.7300475013022663e-05, | |
| "loss": 0.4914, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.08482590629029947, | |
| "grad_norm": 2.2381885051727295, | |
| "learning_rate": 2.3875071764202563e-05, | |
| "loss": 0.5526, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.08711849835219945, | |
| "grad_norm": 3.2347769737243652, | |
| "learning_rate": 2.061073738537635e-05, | |
| "loss": 0.5064, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.08941109041409945, | |
| "grad_norm": 2.4437272548675537, | |
| "learning_rate": 1.7527597583490822e-05, | |
| "loss": 0.4313, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.08941109041409945, | |
| "eval_loss": 0.5332568287849426, | |
| "eval_runtime": 107.5323, | |
| "eval_samples_per_second": 6.835, | |
| "eval_steps_per_second": 3.422, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.09170368247599943, | |
| "grad_norm": 2.3256375789642334, | |
| "learning_rate": 1.4644660940672627e-05, | |
| "loss": 0.4753, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.09399627453789941, | |
| "grad_norm": 4.582269668579102, | |
| "learning_rate": 1.1979701719998453e-05, | |
| "loss": 0.4499, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.09628886659979939, | |
| "grad_norm": 3.641162157058716, | |
| "learning_rate": 9.549150281252633e-06, | |
| "loss": 0.5924, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.09858145866169939, | |
| "grad_norm": 3.2755634784698486, | |
| "learning_rate": 7.367991782295391e-06, | |
| "loss": 0.5981, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.10087405072359937, | |
| "grad_norm": 7.252218246459961, | |
| "learning_rate": 5.449673790581611e-06, | |
| "loss": 0.4751, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.10316664278549935, | |
| "grad_norm": 2.6913466453552246, | |
| "learning_rate": 3.8060233744356633e-06, | |
| "loss": 0.5011, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.10545923484739934, | |
| "grad_norm": 2.337965726852417, | |
| "learning_rate": 2.4471741852423237e-06, | |
| "loss": 0.3562, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.10775182690929933, | |
| "grad_norm": 3.0334508419036865, | |
| "learning_rate": 1.3815039801161721e-06, | |
| "loss": 0.5612, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.1100444189711993, | |
| "grad_norm": 2.4583511352539062, | |
| "learning_rate": 6.15582970243117e-07, | |
| "loss": 0.4611, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.1123370110330993, | |
| "grad_norm": 3.1281447410583496, | |
| "learning_rate": 1.5413331334360182e-07, | |
| "loss": 0.5891, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.11462960309499928, | |
| "grad_norm": 3.4289255142211914, | |
| "learning_rate": 0.0, | |
| "loss": 0.4421, | |
| "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": 2.967549134241792e+17, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |