| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.06436931079323797, | |
| "eval_steps": 9, | |
| "global_step": 99, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0006501950585175553, | |
| "eval_loss": 1.4013111591339111, | |
| "eval_runtime": 233.1503, | |
| "eval_samples_per_second": 11.113, | |
| "eval_steps_per_second": 1.39, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0019505851755526658, | |
| "grad_norm": 1.3066335916519165, | |
| "learning_rate": 3e-05, | |
| "loss": 5.9118, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0039011703511053317, | |
| "grad_norm": 0.6246317625045776, | |
| "learning_rate": 6e-05, | |
| "loss": 5.554, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.005851755526657998, | |
| "grad_norm": 0.7653908729553223, | |
| "learning_rate": 9e-05, | |
| "loss": 5.4893, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.005851755526657998, | |
| "eval_loss": 1.3770792484283447, | |
| "eval_runtime": 234.536, | |
| "eval_samples_per_second": 11.047, | |
| "eval_steps_per_second": 1.381, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.007802340702210663, | |
| "grad_norm": 1.011956810951233, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 5.5717, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.00975292587776333, | |
| "grad_norm": 1.2576535940170288, | |
| "learning_rate": 9.924038765061042e-05, | |
| "loss": 5.2914, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.011703511053315995, | |
| "grad_norm": 1.0005744695663452, | |
| "learning_rate": 9.806308479691595e-05, | |
| "loss": 4.9537, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.011703511053315995, | |
| "eval_loss": 1.255346417427063, | |
| "eval_runtime": 234.5374, | |
| "eval_samples_per_second": 11.047, | |
| "eval_steps_per_second": 1.381, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.013654096228868661, | |
| "grad_norm": 1.09691321849823, | |
| "learning_rate": 9.635919272833938e-05, | |
| "loss": 4.9666, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.015604681404421327, | |
| "grad_norm": 1.3464233875274658, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 4.8085, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.017555266579973992, | |
| "grad_norm": 1.1300208568572998, | |
| "learning_rate": 9.145187862775209e-05, | |
| "loss": 4.8297, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.017555266579973992, | |
| "eval_loss": 1.1976916790008545, | |
| "eval_runtime": 234.5163, | |
| "eval_samples_per_second": 11.048, | |
| "eval_steps_per_second": 1.382, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.01950585175552666, | |
| "grad_norm": 1.766040325164795, | |
| "learning_rate": 8.83022221559489e-05, | |
| "loss": 4.6738, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.021456436931079324, | |
| "grad_norm": 1.0542855262756348, | |
| "learning_rate": 8.473291852294987e-05, | |
| "loss": 4.7766, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.02340702210663199, | |
| "grad_norm": 0.9637995958328247, | |
| "learning_rate": 8.07830737662829e-05, | |
| "loss": 4.5178, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.02340702210663199, | |
| "eval_loss": 1.1739193201065063, | |
| "eval_runtime": 234.5684, | |
| "eval_samples_per_second": 11.046, | |
| "eval_steps_per_second": 1.381, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.025357607282184655, | |
| "grad_norm": 0.858078122138977, | |
| "learning_rate": 7.649596321166024e-05, | |
| "loss": 4.6663, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.027308192457737322, | |
| "grad_norm": 0.7847902774810791, | |
| "learning_rate": 7.191855733945387e-05, | |
| "loss": 4.5735, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.029258777633289986, | |
| "grad_norm": 1.3613299131393433, | |
| "learning_rate": 6.710100716628344e-05, | |
| "loss": 4.6346, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.029258777633289986, | |
| "eval_loss": 1.1611473560333252, | |
| "eval_runtime": 234.5632, | |
| "eval_samples_per_second": 11.046, | |
| "eval_steps_per_second": 1.381, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.031209362808842653, | |
| "grad_norm": 1.0278661251068115, | |
| "learning_rate": 6.209609477998338e-05, | |
| "loss": 4.7762, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.03315994798439532, | |
| "grad_norm": 0.9808676838874817, | |
| "learning_rate": 5.695865504800327e-05, | |
| "loss": 4.5895, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.035110533159947985, | |
| "grad_norm": 0.8073239326477051, | |
| "learning_rate": 5.174497483512506e-05, | |
| "loss": 4.7345, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.035110533159947985, | |
| "eval_loss": 1.152621865272522, | |
| "eval_runtime": 234.5525, | |
| "eval_samples_per_second": 11.047, | |
| "eval_steps_per_second": 1.381, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.03706111833550065, | |
| "grad_norm": 0.8657036423683167, | |
| "learning_rate": 4.6512176312793736e-05, | |
| "loss": 4.8709, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.03901170351105332, | |
| "grad_norm": 0.963033139705658, | |
| "learning_rate": 4.131759111665349e-05, | |
| "loss": 4.6236, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.04096228868660598, | |
| "grad_norm": 1.1742784976959229, | |
| "learning_rate": 3.6218132209150045e-05, | |
| "loss": 4.742, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.04096228868660598, | |
| "eval_loss": 1.1466325521469116, | |
| "eval_runtime": 234.5512, | |
| "eval_samples_per_second": 11.047, | |
| "eval_steps_per_second": 1.381, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.04291287386215865, | |
| "grad_norm": 0.9039548635482788, | |
| "learning_rate": 3.12696703292044e-05, | |
| "loss": 4.6048, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.044863459037711315, | |
| "grad_norm": 0.9454875588417053, | |
| "learning_rate": 2.6526421860705473e-05, | |
| "loss": 4.46, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.04681404421326398, | |
| "grad_norm": 1.141798973083496, | |
| "learning_rate": 2.2040354826462668e-05, | |
| "loss": 4.5007, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.04681404421326398, | |
| "eval_loss": 1.1424474716186523, | |
| "eval_runtime": 234.554, | |
| "eval_samples_per_second": 11.046, | |
| "eval_steps_per_second": 1.381, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.04876462938881664, | |
| "grad_norm": 1.1053383350372314, | |
| "learning_rate": 1.7860619515673033e-05, | |
| "loss": 4.4929, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.05071521456436931, | |
| "grad_norm": 1.6175845861434937, | |
| "learning_rate": 1.4033009983067452e-05, | |
| "loss": 4.5689, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.05266579973992198, | |
| "grad_norm": 1.019932508468628, | |
| "learning_rate": 1.0599462319663905e-05, | |
| "loss": 4.4185, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.05266579973992198, | |
| "eval_loss": 1.1404403448104858, | |
| "eval_runtime": 234.6138, | |
| "eval_samples_per_second": 11.044, | |
| "eval_steps_per_second": 1.381, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.054616384915474644, | |
| "grad_norm": 1.0339871644973755, | |
| "learning_rate": 7.597595192178702e-06, | |
| "loss": 4.4122, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.056566970091027305, | |
| "grad_norm": 1.2999656200408936, | |
| "learning_rate": 5.060297685041659e-06, | |
| "loss": 4.9167, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.05851755526657997, | |
| "grad_norm": 0.8609194159507751, | |
| "learning_rate": 3.0153689607045845e-06, | |
| "loss": 4.6119, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.05851755526657997, | |
| "eval_loss": 1.1394809484481812, | |
| "eval_runtime": 234.4733, | |
| "eval_samples_per_second": 11.05, | |
| "eval_steps_per_second": 1.382, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.06046814044213264, | |
| "grad_norm": 1.0937117338180542, | |
| "learning_rate": 1.4852136862001764e-06, | |
| "loss": 4.6659, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.06241872561768531, | |
| "grad_norm": 1.2036259174346924, | |
| "learning_rate": 4.865965629214819e-07, | |
| "loss": 4.7506, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.06436931079323797, | |
| "grad_norm": 2.0296738147735596, | |
| "learning_rate": 3.04586490452119e-08, | |
| "loss": 4.7603, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.06436931079323797, | |
| "eval_loss": 1.1393488645553589, | |
| "eval_runtime": 234.5356, | |
| "eval_samples_per_second": 11.047, | |
| "eval_steps_per_second": 1.381, | |
| "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": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.293859995821015e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |