| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.2618296529968454, | |
| "eval_steps": 9, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.012618296529968454, | |
| "eval_loss": 1.88910710811615, | |
| "eval_runtime": 2.2136, | |
| "eval_samples_per_second": 60.535, | |
| "eval_steps_per_second": 7.68, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.03785488958990536, | |
| "grad_norm": 0.7511712908744812, | |
| "learning_rate": 3e-05, | |
| "loss": 1.8139, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.07570977917981073, | |
| "grad_norm": 0.8463343381881714, | |
| "learning_rate": 6e-05, | |
| "loss": 1.9489, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.11356466876971609, | |
| "grad_norm": 0.9100675582885742, | |
| "learning_rate": 9e-05, | |
| "loss": 1.7745, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.11356466876971609, | |
| "eval_loss": 1.8236573934555054, | |
| "eval_runtime": 2.232, | |
| "eval_samples_per_second": 60.036, | |
| "eval_steps_per_second": 7.617, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.15141955835962145, | |
| "grad_norm": 0.6830158233642578, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 1.7182, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.1892744479495268, | |
| "grad_norm": 0.7045711278915405, | |
| "learning_rate": 9.924038765061042e-05, | |
| "loss": 1.695, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.22712933753943218, | |
| "grad_norm": 0.6682664752006531, | |
| "learning_rate": 9.806308479691595e-05, | |
| "loss": 1.5894, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.22712933753943218, | |
| "eval_loss": 1.7477922439575195, | |
| "eval_runtime": 2.2514, | |
| "eval_samples_per_second": 59.52, | |
| "eval_steps_per_second": 7.551, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.26498422712933756, | |
| "grad_norm": 0.6020422577857971, | |
| "learning_rate": 9.635919272833938e-05, | |
| "loss": 1.6764, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.3028391167192429, | |
| "grad_norm": 0.8476335406303406, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 1.6508, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.34069400630914826, | |
| "grad_norm": 0.7293965220451355, | |
| "learning_rate": 9.145187862775209e-05, | |
| "loss": 1.4719, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.34069400630914826, | |
| "eval_loss": 1.7059693336486816, | |
| "eval_runtime": 2.2316, | |
| "eval_samples_per_second": 60.048, | |
| "eval_steps_per_second": 7.618, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.3785488958990536, | |
| "grad_norm": 0.6286596059799194, | |
| "learning_rate": 8.83022221559489e-05, | |
| "loss": 1.7276, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.416403785488959, | |
| "grad_norm": 0.8441144227981567, | |
| "learning_rate": 8.473291852294987e-05, | |
| "loss": 1.8129, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.45425867507886436, | |
| "grad_norm": 0.6982610821723938, | |
| "learning_rate": 8.07830737662829e-05, | |
| "loss": 1.8098, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.45425867507886436, | |
| "eval_loss": 1.6810848712921143, | |
| "eval_runtime": 2.2408, | |
| "eval_samples_per_second": 59.8, | |
| "eval_steps_per_second": 7.587, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.4921135646687697, | |
| "grad_norm": 0.704291045665741, | |
| "learning_rate": 7.649596321166024e-05, | |
| "loss": 1.5021, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.5299684542586751, | |
| "grad_norm": 0.725155234336853, | |
| "learning_rate": 7.191855733945387e-05, | |
| "loss": 1.712, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.5678233438485805, | |
| "grad_norm": 0.6382411122322083, | |
| "learning_rate": 6.710100716628344e-05, | |
| "loss": 1.587, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.5678233438485805, | |
| "eval_loss": 1.6629996299743652, | |
| "eval_runtime": 2.2344, | |
| "eval_samples_per_second": 59.971, | |
| "eval_steps_per_second": 7.608, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.6056782334384858, | |
| "grad_norm": 0.717880368232727, | |
| "learning_rate": 6.209609477998338e-05, | |
| "loss": 1.6268, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.6435331230283912, | |
| "grad_norm": 0.688407301902771, | |
| "learning_rate": 5.695865504800327e-05, | |
| "loss": 1.7635, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.6813880126182965, | |
| "grad_norm": 0.7644420266151428, | |
| "learning_rate": 5.174497483512506e-05, | |
| "loss": 1.4975, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.6813880126182965, | |
| "eval_loss": 1.6497859954833984, | |
| "eval_runtime": 2.2418, | |
| "eval_samples_per_second": 59.772, | |
| "eval_steps_per_second": 7.583, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.7192429022082019, | |
| "grad_norm": 0.7091464400291443, | |
| "learning_rate": 4.6512176312793736e-05, | |
| "loss": 1.7363, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.7570977917981072, | |
| "grad_norm": 0.7677245140075684, | |
| "learning_rate": 4.131759111665349e-05, | |
| "loss": 1.4822, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.7949526813880127, | |
| "grad_norm": 0.6542797088623047, | |
| "learning_rate": 3.6218132209150045e-05, | |
| "loss": 1.5378, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.7949526813880127, | |
| "eval_loss": 1.6406694650650024, | |
| "eval_runtime": 2.2416, | |
| "eval_samples_per_second": 59.778, | |
| "eval_steps_per_second": 7.584, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.832807570977918, | |
| "grad_norm": 0.718596875667572, | |
| "learning_rate": 3.12696703292044e-05, | |
| "loss": 1.5353, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.8706624605678234, | |
| "grad_norm": 0.6250401735305786, | |
| "learning_rate": 2.6526421860705473e-05, | |
| "loss": 1.5808, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.9085173501577287, | |
| "grad_norm": 0.7185081243515015, | |
| "learning_rate": 2.2040354826462668e-05, | |
| "loss": 1.8177, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.9085173501577287, | |
| "eval_loss": 1.6348599195480347, | |
| "eval_runtime": 2.2303, | |
| "eval_samples_per_second": 60.082, | |
| "eval_steps_per_second": 7.622, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.9463722397476341, | |
| "grad_norm": 0.8165435194969177, | |
| "learning_rate": 1.7860619515673033e-05, | |
| "loss": 1.6547, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.9842271293375394, | |
| "grad_norm": 0.7258487343788147, | |
| "learning_rate": 1.4033009983067452e-05, | |
| "loss": 1.5405, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 1.0220820189274449, | |
| "grad_norm": 0.6014052629470825, | |
| "learning_rate": 1.0599462319663905e-05, | |
| "loss": 1.8766, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 1.0220820189274449, | |
| "eval_loss": 1.6316182613372803, | |
| "eval_runtime": 2.2331, | |
| "eval_samples_per_second": 60.007, | |
| "eval_steps_per_second": 7.613, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 1.0599369085173502, | |
| "grad_norm": 0.8883140683174133, | |
| "learning_rate": 7.597595192178702e-06, | |
| "loss": 1.7513, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 1.0977917981072556, | |
| "grad_norm": 0.7024074196815491, | |
| "learning_rate": 5.060297685041659e-06, | |
| "loss": 1.3722, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 1.135646687697161, | |
| "grad_norm": 0.8499208092689514, | |
| "learning_rate": 3.0153689607045845e-06, | |
| "loss": 1.4446, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.135646687697161, | |
| "eval_loss": 1.630479097366333, | |
| "eval_runtime": 2.2459, | |
| "eval_samples_per_second": 59.663, | |
| "eval_steps_per_second": 7.569, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.1735015772870663, | |
| "grad_norm": 0.800684928894043, | |
| "learning_rate": 1.4852136862001764e-06, | |
| "loss": 1.6495, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 1.2113564668769716, | |
| "grad_norm": 0.615403950214386, | |
| "learning_rate": 4.865965629214819e-07, | |
| "loss": 1.5472, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 1.249211356466877, | |
| "grad_norm": 0.7376525402069092, | |
| "learning_rate": 3.04586490452119e-08, | |
| "loss": 1.3958, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 1.249211356466877, | |
| "eval_loss": 1.6302069425582886, | |
| "eval_runtime": 2.2327, | |
| "eval_samples_per_second": 60.017, | |
| "eval_steps_per_second": 7.614, | |
| "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": 7874172754329600.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |