| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.172514619883041, | |
| "eval_steps": 9, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.011695906432748537, | |
| "eval_loss": 1.4764314889907837, | |
| "eval_runtime": 13.5735, | |
| "eval_samples_per_second": 10.609, | |
| "eval_steps_per_second": 1.326, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.03508771929824561, | |
| "grad_norm": 0.47823527455329895, | |
| "learning_rate": 3e-05, | |
| "loss": 5.7719, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.07017543859649122, | |
| "grad_norm": 0.6247535943984985, | |
| "learning_rate": 6e-05, | |
| "loss": 5.836, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.10526315789473684, | |
| "grad_norm": 0.5008035898208618, | |
| "learning_rate": 9e-05, | |
| "loss": 5.5911, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.10526315789473684, | |
| "eval_loss": 1.450061559677124, | |
| "eval_runtime": 13.8272, | |
| "eval_samples_per_second": 10.414, | |
| "eval_steps_per_second": 1.302, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.14035087719298245, | |
| "grad_norm": 0.5038430094718933, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 5.6881, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.17543859649122806, | |
| "grad_norm": 0.6504800319671631, | |
| "learning_rate": 9.924038765061042e-05, | |
| "loss": 5.8583, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.21052631578947367, | |
| "grad_norm": 0.47255322337150574, | |
| "learning_rate": 9.806308479691595e-05, | |
| "loss": 5.4709, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.21052631578947367, | |
| "eval_loss": 1.407480001449585, | |
| "eval_runtime": 13.8137, | |
| "eval_samples_per_second": 10.424, | |
| "eval_steps_per_second": 1.303, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.24561403508771928, | |
| "grad_norm": 0.393683522939682, | |
| "learning_rate": 9.635919272833938e-05, | |
| "loss": 5.346, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.2807017543859649, | |
| "grad_norm": 0.40169697999954224, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 5.3727, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.3157894736842105, | |
| "grad_norm": 0.37064439058303833, | |
| "learning_rate": 9.145187862775209e-05, | |
| "loss": 5.2459, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.3157894736842105, | |
| "eval_loss": 1.3822287321090698, | |
| "eval_runtime": 13.7588, | |
| "eval_samples_per_second": 10.466, | |
| "eval_steps_per_second": 1.308, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.3508771929824561, | |
| "grad_norm": 0.44399482011795044, | |
| "learning_rate": 8.83022221559489e-05, | |
| "loss": 5.6151, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.38596491228070173, | |
| "grad_norm": 0.4025361239910126, | |
| "learning_rate": 8.473291852294987e-05, | |
| "loss": 5.1451, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.42105263157894735, | |
| "grad_norm": 0.3941345810890198, | |
| "learning_rate": 8.07830737662829e-05, | |
| "loss": 4.9181, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.42105263157894735, | |
| "eval_loss": 1.3653277158737183, | |
| "eval_runtime": 13.7475, | |
| "eval_samples_per_second": 10.475, | |
| "eval_steps_per_second": 1.309, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.45614035087719296, | |
| "grad_norm": 0.38964855670928955, | |
| "learning_rate": 7.649596321166024e-05, | |
| "loss": 5.4978, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.49122807017543857, | |
| "grad_norm": 0.6319313049316406, | |
| "learning_rate": 7.191855733945387e-05, | |
| "loss": 5.203, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.5263157894736842, | |
| "grad_norm": 0.43766701221466064, | |
| "learning_rate": 6.710100716628344e-05, | |
| "loss": 5.3021, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.5263157894736842, | |
| "eval_loss": 1.351959466934204, | |
| "eval_runtime": 13.7383, | |
| "eval_samples_per_second": 10.482, | |
| "eval_steps_per_second": 1.31, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.5614035087719298, | |
| "grad_norm": 0.40817752480506897, | |
| "learning_rate": 6.209609477998338e-05, | |
| "loss": 5.3959, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.5964912280701754, | |
| "grad_norm": 0.5033959746360779, | |
| "learning_rate": 5.695865504800327e-05, | |
| "loss": 5.1644, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.631578947368421, | |
| "grad_norm": 0.44770410656929016, | |
| "learning_rate": 5.174497483512506e-05, | |
| "loss": 5.485, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.631578947368421, | |
| "eval_loss": 1.3431975841522217, | |
| "eval_runtime": 13.7444, | |
| "eval_samples_per_second": 10.477, | |
| "eval_steps_per_second": 1.31, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.6666666666666666, | |
| "grad_norm": 0.4134134352207184, | |
| "learning_rate": 4.6512176312793736e-05, | |
| "loss": 5.1448, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.7017543859649122, | |
| "grad_norm": 0.40665504336357117, | |
| "learning_rate": 4.131759111665349e-05, | |
| "loss": 4.8621, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.7368421052631579, | |
| "grad_norm": 0.47186246514320374, | |
| "learning_rate": 3.6218132209150045e-05, | |
| "loss": 4.9763, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.7368421052631579, | |
| "eval_loss": 1.3367018699645996, | |
| "eval_runtime": 13.7427, | |
| "eval_samples_per_second": 10.478, | |
| "eval_steps_per_second": 1.31, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.7719298245614035, | |
| "grad_norm": 0.4636601209640503, | |
| "learning_rate": 3.12696703292044e-05, | |
| "loss": 4.8446, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.8070175438596491, | |
| "grad_norm": 0.5160920023918152, | |
| "learning_rate": 2.6526421860705473e-05, | |
| "loss": 5.3747, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.8421052631578947, | |
| "grad_norm": 0.46744614839553833, | |
| "learning_rate": 2.2040354826462668e-05, | |
| "loss": 5.1907, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.8421052631578947, | |
| "eval_loss": 1.3317314386367798, | |
| "eval_runtime": 13.7478, | |
| "eval_samples_per_second": 10.474, | |
| "eval_steps_per_second": 1.309, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.8771929824561403, | |
| "grad_norm": 0.4592737555503845, | |
| "learning_rate": 1.7860619515673033e-05, | |
| "loss": 5.4021, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.9122807017543859, | |
| "grad_norm": 0.46895694732666016, | |
| "learning_rate": 1.4033009983067452e-05, | |
| "loss": 5.2216, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.9473684210526315, | |
| "grad_norm": 0.49175575375556946, | |
| "learning_rate": 1.0599462319663905e-05, | |
| "loss": 5.1936, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.9473684210526315, | |
| "eval_loss": 1.3289103507995605, | |
| "eval_runtime": 13.7459, | |
| "eval_samples_per_second": 10.476, | |
| "eval_steps_per_second": 1.309, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.9824561403508771, | |
| "grad_norm": 0.45295849442481995, | |
| "learning_rate": 7.597595192178702e-06, | |
| "loss": 5.2037, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 1.0204678362573099, | |
| "grad_norm": 0.4632893204689026, | |
| "learning_rate": 5.060297685041659e-06, | |
| "loss": 4.9032, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 1.0555555555555556, | |
| "grad_norm": 0.5045644044876099, | |
| "learning_rate": 3.0153689607045845e-06, | |
| "loss": 5.235, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.0555555555555556, | |
| "eval_loss": 1.3276033401489258, | |
| "eval_runtime": 13.751, | |
| "eval_samples_per_second": 10.472, | |
| "eval_steps_per_second": 1.309, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.090643274853801, | |
| "grad_norm": 0.44458574056625366, | |
| "learning_rate": 1.4852136862001764e-06, | |
| "loss": 4.9056, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 1.1257309941520468, | |
| "grad_norm": 0.47579580545425415, | |
| "learning_rate": 4.865965629214819e-07, | |
| "loss": 4.9345, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 1.1608187134502923, | |
| "grad_norm": 0.508492648601532, | |
| "learning_rate": 3.04586490452119e-08, | |
| "loss": 5.1551, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 1.1608187134502923, | |
| "eval_loss": 1.327457308769226, | |
| "eval_runtime": 13.7559, | |
| "eval_samples_per_second": 10.468, | |
| "eval_steps_per_second": 1.309, | |
| "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": 1.47552863256576e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |