| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.0, | |
| "eval_steps": 500, | |
| "global_step": 32, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0625, | |
| "grad_norm": 0.03886242265457615, | |
| "learning_rate": 1.25e-05, | |
| "loss": 0.1926, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.125, | |
| "grad_norm": 0.03867154797319413, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.1964, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.1875, | |
| "grad_norm": 0.03859478229432736, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 0.186, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "grad_norm": 0.04297586914169591, | |
| "learning_rate": 5e-05, | |
| "loss": 0.188, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.3125, | |
| "grad_norm": 0.04661154126462952, | |
| "learning_rate": 4.984280524733107e-05, | |
| "loss": 0.1758, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.375, | |
| "grad_norm": 0.044661750490632465, | |
| "learning_rate": 4.937319780454559e-05, | |
| "loss": 0.1734, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.4375, | |
| "grad_norm": 0.043466066790942615, | |
| "learning_rate": 4.8597083257709194e-05, | |
| "loss": 0.1613, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": 0.03160423221050834, | |
| "learning_rate": 4.752422169756048e-05, | |
| "loss": 0.1559, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.5625, | |
| "grad_norm": 0.0278736495424202, | |
| "learning_rate": 4.6168104980707107e-05, | |
| "loss": 0.1588, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.625, | |
| "grad_norm": 0.033644420390984725, | |
| "learning_rate": 4.454578706170075e-05, | |
| "loss": 0.1585, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.6875, | |
| "grad_norm": 0.045043888693144754, | |
| "learning_rate": 4.267766952966369e-05, | |
| "loss": 0.1584, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "grad_norm": 0.059694975441249985, | |
| "learning_rate": 4.058724504646834e-05, | |
| "loss": 0.1498, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.8125, | |
| "grad_norm": 0.04201384738580901, | |
| "learning_rate": 3.830080191288342e-05, | |
| "loss": 0.1474, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.875, | |
| "grad_norm": 0.03660468512160384, | |
| "learning_rate": 3.5847093477938956e-05, | |
| "loss": 0.1455, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.9375, | |
| "grad_norm": 0.03202574071597613, | |
| "learning_rate": 3.3256976548879184e-05, | |
| "loss": 0.1453, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 0.022553871083231203, | |
| "learning_rate": 3.056302334890786e-05, | |
| "loss": 0.1454, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 1.0625, | |
| "grad_norm": 0.018270438778563842, | |
| "learning_rate": 2.7799111902582696e-05, | |
| "loss": 0.1295, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 1.125, | |
| "grad_norm": 0.018025445019069265, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.1352, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 1.1875, | |
| "grad_norm": 0.01938426125298895, | |
| "learning_rate": 2.2200888097417307e-05, | |
| "loss": 0.1267, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "grad_norm": 0.024835127196697317, | |
| "learning_rate": 1.9436976651092144e-05, | |
| "loss": 0.1213, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.3125, | |
| "grad_norm": 0.02442204305328, | |
| "learning_rate": 1.6743023451120832e-05, | |
| "loss": 0.1324, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 1.375, | |
| "grad_norm": 0.02283960256436579, | |
| "learning_rate": 1.4152906522061048e-05, | |
| "loss": 0.1371, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 1.4375, | |
| "grad_norm": 0.02373187865474639, | |
| "learning_rate": 1.1699198087116589e-05, | |
| "loss": 0.1317, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "grad_norm": 0.018742991552912824, | |
| "learning_rate": 9.412754953531663e-06, | |
| "loss": 0.1261, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 1.5625, | |
| "grad_norm": 0.020956665698211793, | |
| "learning_rate": 7.3223304703363135e-06, | |
| "loss": 0.1314, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 1.625, | |
| "grad_norm": 0.019876581721580355, | |
| "learning_rate": 5.454212938299255e-06, | |
| "loss": 0.129, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 1.6875, | |
| "grad_norm": 0.020417609679245066, | |
| "learning_rate": 3.831895019292897e-06, | |
| "loss": 0.1263, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "grad_norm": 0.01988593677695763, | |
| "learning_rate": 2.475778302439524e-06, | |
| "loss": 0.1338, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 1.8125, | |
| "grad_norm": 0.01911980479645202, | |
| "learning_rate": 1.4029167422908107e-06, | |
| "loss": 0.1307, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 1.875, | |
| "grad_norm": 0.021333837613636705, | |
| "learning_rate": 6.268021954544096e-07, | |
| "loss": 0.134, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.9375, | |
| "grad_norm": 0.023428038302216552, | |
| "learning_rate": 1.571947526689349e-07, | |
| "loss": 0.1251, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 0.020881132010724595, | |
| "learning_rate": 0.0, | |
| "loss": 0.1214, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "step": 32, | |
| "total_flos": 593996940312576.0, | |
| "train_loss": 0.14719508402049541, | |
| "train_runtime": 1046.9269, | |
| "train_samples_per_second": 0.478, | |
| "train_steps_per_second": 0.031 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 32, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 2, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 593996940312576.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |