| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.292604501607717, | |
| "eval_steps": 9, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.012861736334405145, | |
| "eval_loss": 10.379566192626953, | |
| "eval_runtime": 0.2439, | |
| "eval_samples_per_second": 537.151, | |
| "eval_steps_per_second": 69.707, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.03858520900321544, | |
| "grad_norm": 0.035851553082466125, | |
| "learning_rate": 3e-05, | |
| "loss": 10.3768, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.07717041800643087, | |
| "grad_norm": 0.02997441403567791, | |
| "learning_rate": 6e-05, | |
| "loss": 10.3771, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.1157556270096463, | |
| "grad_norm": 0.03328309953212738, | |
| "learning_rate": 9e-05, | |
| "loss": 10.3772, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.1157556270096463, | |
| "eval_loss": 10.379271507263184, | |
| "eval_runtime": 0.2441, | |
| "eval_samples_per_second": 536.679, | |
| "eval_steps_per_second": 69.645, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.15434083601286175, | |
| "grad_norm": 0.03068096563220024, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 10.3742, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.19292604501607716, | |
| "grad_norm": 0.03114703670144081, | |
| "learning_rate": 9.924038765061042e-05, | |
| "loss": 10.3783, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.2315112540192926, | |
| "grad_norm": 0.03824223577976227, | |
| "learning_rate": 9.806308479691595e-05, | |
| "loss": 10.3772, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.2315112540192926, | |
| "eval_loss": 10.378520011901855, | |
| "eval_runtime": 0.2586, | |
| "eval_samples_per_second": 506.574, | |
| "eval_steps_per_second": 65.739, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.27009646302250806, | |
| "grad_norm": 0.03710794448852539, | |
| "learning_rate": 9.635919272833938e-05, | |
| "loss": 10.3762, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.3086816720257235, | |
| "grad_norm": 0.035997722297906876, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 10.3755, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.34726688102893893, | |
| "grad_norm": 0.03549078106880188, | |
| "learning_rate": 9.145187862775209e-05, | |
| "loss": 10.3761, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.34726688102893893, | |
| "eval_loss": 10.377715110778809, | |
| "eval_runtime": 0.2406, | |
| "eval_samples_per_second": 544.381, | |
| "eval_steps_per_second": 70.645, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.3858520900321543, | |
| "grad_norm": 0.0394701287150383, | |
| "learning_rate": 8.83022221559489e-05, | |
| "loss": 10.3771, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.42443729903536975, | |
| "grad_norm": 0.034668490290641785, | |
| "learning_rate": 8.473291852294987e-05, | |
| "loss": 10.3753, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.4630225080385852, | |
| "grad_norm": 0.05240073800086975, | |
| "learning_rate": 8.07830737662829e-05, | |
| "loss": 10.3784, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.4630225080385852, | |
| "eval_loss": 10.37689208984375, | |
| "eval_runtime": 0.246, | |
| "eval_samples_per_second": 532.568, | |
| "eval_steps_per_second": 69.112, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.5016077170418006, | |
| "grad_norm": 0.04496760293841362, | |
| "learning_rate": 7.649596321166024e-05, | |
| "loss": 10.3757, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.5401929260450161, | |
| "grad_norm": 0.05573504418134689, | |
| "learning_rate": 7.191855733945387e-05, | |
| "loss": 10.3767, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.5787781350482315, | |
| "grad_norm": 0.04383615776896477, | |
| "learning_rate": 6.710100716628344e-05, | |
| "loss": 10.3767, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.5787781350482315, | |
| "eval_loss": 10.375993728637695, | |
| "eval_runtime": 0.2399, | |
| "eval_samples_per_second": 546.082, | |
| "eval_steps_per_second": 70.866, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.617363344051447, | |
| "grad_norm": 0.053934141993522644, | |
| "learning_rate": 6.209609477998338e-05, | |
| "loss": 10.3738, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.6559485530546624, | |
| "grad_norm": 0.05161559209227562, | |
| "learning_rate": 5.695865504800327e-05, | |
| "loss": 10.3746, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.6945337620578779, | |
| "grad_norm": 0.0588853619992733, | |
| "learning_rate": 5.174497483512506e-05, | |
| "loss": 10.3733, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.6945337620578779, | |
| "eval_loss": 10.375133514404297, | |
| "eval_runtime": 0.241, | |
| "eval_samples_per_second": 543.574, | |
| "eval_steps_per_second": 70.54, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.7331189710610932, | |
| "grad_norm": 0.060255225747823715, | |
| "learning_rate": 4.6512176312793736e-05, | |
| "loss": 10.3748, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.7717041800643086, | |
| "grad_norm": 0.07191349565982819, | |
| "learning_rate": 4.131759111665349e-05, | |
| "loss": 10.373, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.8102893890675241, | |
| "grad_norm": 0.0634809359908104, | |
| "learning_rate": 3.6218132209150045e-05, | |
| "loss": 10.3714, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.8102893890675241, | |
| "eval_loss": 10.374332427978516, | |
| "eval_runtime": 0.2398, | |
| "eval_samples_per_second": 546.342, | |
| "eval_steps_per_second": 70.899, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.8488745980707395, | |
| "grad_norm": 0.06558206677436829, | |
| "learning_rate": 3.12696703292044e-05, | |
| "loss": 10.3727, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.887459807073955, | |
| "grad_norm": 0.06752001494169235, | |
| "learning_rate": 2.6526421860705473e-05, | |
| "loss": 10.3736, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.9260450160771704, | |
| "grad_norm": 0.06886977702379227, | |
| "learning_rate": 2.2040354826462668e-05, | |
| "loss": 10.3754, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.9260450160771704, | |
| "eval_loss": 10.373759269714355, | |
| "eval_runtime": 0.2468, | |
| "eval_samples_per_second": 530.721, | |
| "eval_steps_per_second": 68.872, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.9646302250803859, | |
| "grad_norm": 0.062006477266550064, | |
| "learning_rate": 1.7860619515673033e-05, | |
| "loss": 10.3717, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 1.0096463022508038, | |
| "grad_norm": 0.12904778122901917, | |
| "learning_rate": 1.4033009983067452e-05, | |
| "loss": 12.9222, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 1.0482315112540193, | |
| "grad_norm": 0.0823853611946106, | |
| "learning_rate": 1.0599462319663905e-05, | |
| "loss": 10.5007, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 1.0482315112540193, | |
| "eval_loss": 10.373393058776855, | |
| "eval_runtime": 0.2576, | |
| "eval_samples_per_second": 508.556, | |
| "eval_steps_per_second": 65.996, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 1.0868167202572347, | |
| "grad_norm": 0.07457450032234192, | |
| "learning_rate": 7.597595192178702e-06, | |
| "loss": 10.245, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 1.1254019292604502, | |
| "grad_norm": 0.07634331285953522, | |
| "learning_rate": 5.060297685041659e-06, | |
| "loss": 10.3788, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 1.1639871382636655, | |
| "grad_norm": 0.05044448748230934, | |
| "learning_rate": 3.0153689607045845e-06, | |
| "loss": 10.4191, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.1639871382636655, | |
| "eval_loss": 10.373234748840332, | |
| "eval_runtime": 0.2371, | |
| "eval_samples_per_second": 552.416, | |
| "eval_steps_per_second": 71.688, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.202572347266881, | |
| "grad_norm": 0.070530004799366, | |
| "learning_rate": 1.4852136862001764e-06, | |
| "loss": 10.4401, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 1.2411575562700965, | |
| "grad_norm": 0.07272198051214218, | |
| "learning_rate": 4.865965629214819e-07, | |
| "loss": 10.3003, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 1.279742765273312, | |
| "grad_norm": 0.0831335186958313, | |
| "learning_rate": 3.04586490452119e-08, | |
| "loss": 10.4156, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 1.279742765273312, | |
| "eval_loss": 10.373204231262207, | |
| "eval_runtime": 0.2509, | |
| "eval_samples_per_second": 522.123, | |
| "eval_steps_per_second": 67.756, | |
| "step": 99 | |
| } | |
| ], | |
| "logging_steps": 3, | |
| "max_steps": 100, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 2, | |
| "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": 10460489318400.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |