| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.5805515239477503, | |
| "eval_steps": 9, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.005805515239477504, | |
| "eval_loss": 0.1446908712387085, | |
| "eval_runtime": 35.8549, | |
| "eval_samples_per_second": 8.088, | |
| "eval_steps_per_second": 1.032, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.01741654571843251, | |
| "grad_norm": 8.589686393737793, | |
| "learning_rate": 3e-05, | |
| "loss": 0.6468, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.03483309143686502, | |
| "grad_norm": 2.196739435195923, | |
| "learning_rate": 6e-05, | |
| "loss": 0.3215, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.05224963715529753, | |
| "grad_norm": 1.0070608854293823, | |
| "learning_rate": 9e-05, | |
| "loss": 0.1349, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.05224963715529753, | |
| "eval_loss": 0.04279467836022377, | |
| "eval_runtime": 36.4605, | |
| "eval_samples_per_second": 7.954, | |
| "eval_steps_per_second": 1.015, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.06966618287373004, | |
| "grad_norm": 1.272331953048706, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 0.0889, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.08708272859216255, | |
| "grad_norm": 0.8801549077033997, | |
| "learning_rate": 9.924038765061042e-05, | |
| "loss": 0.1656, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.10449927431059507, | |
| "grad_norm": 1.045451045036316, | |
| "learning_rate": 9.806308479691595e-05, | |
| "loss": 0.0636, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.10449927431059507, | |
| "eval_loss": 0.031937576830387115, | |
| "eval_runtime": 36.4522, | |
| "eval_samples_per_second": 7.956, | |
| "eval_steps_per_second": 1.015, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.12191582002902758, | |
| "grad_norm": 0.48578259348869324, | |
| "learning_rate": 9.635919272833938e-05, | |
| "loss": 0.1489, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.13933236574746008, | |
| "grad_norm": 0.5960637927055359, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 0.0821, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.1567489114658926, | |
| "grad_norm": 0.6202490925788879, | |
| "learning_rate": 9.145187862775209e-05, | |
| "loss": 0.1026, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.1567489114658926, | |
| "eval_loss": 0.026807833462953568, | |
| "eval_runtime": 36.4704, | |
| "eval_samples_per_second": 7.952, | |
| "eval_steps_per_second": 1.015, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.1741654571843251, | |
| "grad_norm": 0.47882765531539917, | |
| "learning_rate": 8.83022221559489e-05, | |
| "loss": 0.0797, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.19158200290275762, | |
| "grad_norm": 0.5080346465110779, | |
| "learning_rate": 8.473291852294987e-05, | |
| "loss": 0.0137, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.20899854862119013, | |
| "grad_norm": 0.8254001140594482, | |
| "learning_rate": 8.07830737662829e-05, | |
| "loss": 0.0772, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.20899854862119013, | |
| "eval_loss": 0.02542661316692829, | |
| "eval_runtime": 36.4687, | |
| "eval_samples_per_second": 7.952, | |
| "eval_steps_per_second": 1.015, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.22641509433962265, | |
| "grad_norm": 0.47083836793899536, | |
| "learning_rate": 7.649596321166024e-05, | |
| "loss": 0.0673, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.24383164005805516, | |
| "grad_norm": 0.5346234440803528, | |
| "learning_rate": 7.191855733945387e-05, | |
| "loss": 0.0242, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.2612481857764877, | |
| "grad_norm": 0.42815378308296204, | |
| "learning_rate": 6.710100716628344e-05, | |
| "loss": 0.0727, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.2612481857764877, | |
| "eval_loss": 0.02381318248808384, | |
| "eval_runtime": 36.4961, | |
| "eval_samples_per_second": 7.946, | |
| "eval_steps_per_second": 1.014, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.27866473149492016, | |
| "grad_norm": 0.5860163569450378, | |
| "learning_rate": 6.209609477998338e-05, | |
| "loss": 0.0393, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.2960812772133527, | |
| "grad_norm": 0.3074454069137573, | |
| "learning_rate": 5.695865504800327e-05, | |
| "loss": 0.0321, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.3134978229317852, | |
| "grad_norm": 0.23264048993587494, | |
| "learning_rate": 5.174497483512506e-05, | |
| "loss": 0.0234, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.3134978229317852, | |
| "eval_loss": 0.022688375785946846, | |
| "eval_runtime": 36.4879, | |
| "eval_samples_per_second": 7.948, | |
| "eval_steps_per_second": 1.014, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.3309143686502177, | |
| "grad_norm": 0.21743811666965485, | |
| "learning_rate": 4.6512176312793736e-05, | |
| "loss": 0.0089, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.3483309143686502, | |
| "grad_norm": 0.5799148678779602, | |
| "learning_rate": 4.131759111665349e-05, | |
| "loss": 0.0938, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.36574746008708275, | |
| "grad_norm": 0.4093296527862549, | |
| "learning_rate": 3.6218132209150045e-05, | |
| "loss": 0.1154, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.36574746008708275, | |
| "eval_loss": 0.021804720163345337, | |
| "eval_runtime": 36.4988, | |
| "eval_samples_per_second": 7.945, | |
| "eval_steps_per_second": 1.014, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.38316400580551524, | |
| "grad_norm": 0.29445552825927734, | |
| "learning_rate": 3.12696703292044e-05, | |
| "loss": 0.0195, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.4005805515239477, | |
| "grad_norm": 0.19134579598903656, | |
| "learning_rate": 2.6526421860705473e-05, | |
| "loss": 0.0057, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.41799709724238027, | |
| "grad_norm": 0.5365853905677795, | |
| "learning_rate": 2.2040354826462668e-05, | |
| "loss": 0.0612, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.41799709724238027, | |
| "eval_loss": 0.02147766947746277, | |
| "eval_runtime": 36.4726, | |
| "eval_samples_per_second": 7.951, | |
| "eval_steps_per_second": 1.014, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.43541364296081275, | |
| "grad_norm": 0.14221055805683136, | |
| "learning_rate": 1.7860619515673033e-05, | |
| "loss": 0.0321, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.4528301886792453, | |
| "grad_norm": 0.52949059009552, | |
| "learning_rate": 1.4033009983067452e-05, | |
| "loss": 0.0412, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.4702467343976778, | |
| "grad_norm": 0.6387705206871033, | |
| "learning_rate": 1.0599462319663905e-05, | |
| "loss": 0.0287, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.4702467343976778, | |
| "eval_loss": 0.021191399544477463, | |
| "eval_runtime": 36.4534, | |
| "eval_samples_per_second": 7.955, | |
| "eval_steps_per_second": 1.015, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.4876632801161103, | |
| "grad_norm": 0.5022262334823608, | |
| "learning_rate": 7.597595192178702e-06, | |
| "loss": 0.1063, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.5050798258345428, | |
| "grad_norm": 0.40936022996902466, | |
| "learning_rate": 5.060297685041659e-06, | |
| "loss": 0.0505, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.5224963715529753, | |
| "grad_norm": 0.6321708559989929, | |
| "learning_rate": 3.0153689607045845e-06, | |
| "loss": 0.0997, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.5224963715529753, | |
| "eval_loss": 0.020932814106345177, | |
| "eval_runtime": 36.4786, | |
| "eval_samples_per_second": 7.95, | |
| "eval_steps_per_second": 1.014, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.5399129172714079, | |
| "grad_norm": 0.890493631362915, | |
| "learning_rate": 1.4852136862001764e-06, | |
| "loss": 0.0538, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.5573294629898403, | |
| "grad_norm": 0.6564047336578369, | |
| "learning_rate": 4.865965629214819e-07, | |
| "loss": 0.0611, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.5747460087082729, | |
| "grad_norm": 0.3723383843898773, | |
| "learning_rate": 3.04586490452119e-08, | |
| "loss": 0.0388, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.5747460087082729, | |
| "eval_loss": 0.020911818370223045, | |
| "eval_runtime": 36.4731, | |
| "eval_samples_per_second": 7.951, | |
| "eval_steps_per_second": 1.014, | |
| "step": 99 | |
| } | |
| ], | |
| "logging_steps": 3, | |
| "max_steps": 100, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 9, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2.383609000529756e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |