| { | |
| "best_global_step": null, | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.0, | |
| "eval_steps": 51, | |
| "global_step": 204, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.05889570552147239, | |
| "grad_norm": 0.17814360558986664, | |
| "learning_rate": 9.803921568627451e-06, | |
| "loss": 0.1429, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.11779141104294479, | |
| "grad_norm": 0.19341157376766205, | |
| "learning_rate": 2.1568627450980395e-05, | |
| "loss": 0.1421, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.17668711656441718, | |
| "grad_norm": 0.21557286381721497, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 0.1137, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.23558282208588957, | |
| "grad_norm": 0.0923461839556694, | |
| "learning_rate": 4.5098039215686275e-05, | |
| "loss": 0.0903, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.294478527607362, | |
| "grad_norm": 0.0747547447681427, | |
| "learning_rate": 5.6862745098039215e-05, | |
| "loss": 0.0541, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.35337423312883437, | |
| "grad_norm": 0.06449007987976074, | |
| "learning_rate": 6.862745098039216e-05, | |
| "loss": 0.056, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.41226993865030676, | |
| "grad_norm": 0.06695199757814407, | |
| "learning_rate": 8.039215686274511e-05, | |
| "loss": 0.0378, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.47116564417177914, | |
| "grad_norm": 0.057652052491903305, | |
| "learning_rate": 9.215686274509804e-05, | |
| "loss": 0.0333, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.5006134969325153, | |
| "eval_loss": 0.04454955831170082, | |
| "eval_runtime": 99.1976, | |
| "eval_samples_per_second": 0.917, | |
| "eval_steps_per_second": 0.917, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.5300613496932516, | |
| "grad_norm": 0.07416237890720367, | |
| "learning_rate": 9.99952117026961e-05, | |
| "loss": 0.0317, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.588957055214724, | |
| "grad_norm": 0.0759715884923935, | |
| "learning_rate": 9.992340558396519e-05, | |
| "loss": 0.0333, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.6478527607361964, | |
| "grad_norm": 0.06889687478542328, | |
| "learning_rate": 9.976555313435489e-05, | |
| "loss": 0.0265, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.7067484662576687, | |
| "grad_norm": 0.06610561907291412, | |
| "learning_rate": 9.952192642312712e-05, | |
| "loss": 0.0267, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.7656441717791411, | |
| "grad_norm": 0.05757423862814903, | |
| "learning_rate": 9.919294535722452e-05, | |
| "loss": 0.0276, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.8245398773006135, | |
| "grad_norm": 0.05512451380491257, | |
| "learning_rate": 9.877917695753274e-05, | |
| "loss": 0.0224, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.8834355828220859, | |
| "grad_norm": 0.06681577116250992, | |
| "learning_rate": 9.828133438158205e-05, | |
| "loss": 0.02, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.9423312883435583, | |
| "grad_norm": 0.07138796895742416, | |
| "learning_rate": 9.770027569437253e-05, | |
| "loss": 0.0263, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 0.06135529652237892, | |
| "learning_rate": 9.703700238944158e-05, | |
| "loss": 0.0189, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.027291039004921913, | |
| "eval_runtime": 90.02, | |
| "eval_samples_per_second": 1.011, | |
| "eval_steps_per_second": 1.011, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 1.0588957055214725, | |
| "grad_norm": 0.09823279082775116, | |
| "learning_rate": 9.629265766272292e-05, | |
| "loss": 0.0158, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 1.1177914110429448, | |
| "grad_norm": 0.05707687512040138, | |
| "learning_rate": 9.54685244421718e-05, | |
| "loss": 0.017, | |
| "step": 114 | |
| }, | |
| { | |
| "epoch": 1.1766871165644173, | |
| "grad_norm": 0.06987980008125305, | |
| "learning_rate": 9.456602317655275e-05, | |
| "loss": 0.0169, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.2355828220858895, | |
| "grad_norm": 0.06296130269765854, | |
| "learning_rate": 9.358670938720113e-05, | |
| "loss": 0.0162, | |
| "step": 126 | |
| }, | |
| { | |
| "epoch": 1.294478527607362, | |
| "grad_norm": 0.059489328414201736, | |
| "learning_rate": 9.253227098697803e-05, | |
| "loss": 0.0166, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 1.3533742331288343, | |
| "grad_norm": 0.06373760849237442, | |
| "learning_rate": 9.140452537103942e-05, | |
| "loss": 0.0188, | |
| "step": 138 | |
| }, | |
| { | |
| "epoch": 1.4122699386503068, | |
| "grad_norm": 0.06238365173339844, | |
| "learning_rate": 9.020541628443395e-05, | |
| "loss": 0.0161, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 1.471165644171779, | |
| "grad_norm": 0.071172334253788, | |
| "learning_rate": 8.893701047192833e-05, | |
| "loss": 0.0196, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.5006134969325153, | |
| "eval_loss": 0.024013910442590714, | |
| "eval_runtime": 90.4486, | |
| "eval_samples_per_second": 1.006, | |
| "eval_steps_per_second": 1.006, | |
| "step": 153 | |
| }, | |
| { | |
| "epoch": 1.5300613496932516, | |
| "grad_norm": 0.06220275163650513, | |
| "learning_rate": 8.760149411583437e-05, | |
| "loss": 0.0185, | |
| "step": 156 | |
| }, | |
| { | |
| "epoch": 1.588957055214724, | |
| "grad_norm": 0.06537278741598129, | |
| "learning_rate": 8.620116906797739e-05, | |
| "loss": 0.0227, | |
| "step": 162 | |
| }, | |
| { | |
| "epoch": 1.6478527607361964, | |
| "grad_norm": 0.04823810234665871, | |
| "learning_rate": 8.473844888230065e-05, | |
| "loss": 0.019, | |
| "step": 168 | |
| }, | |
| { | |
| "epoch": 1.7067484662576686, | |
| "grad_norm": 0.049262188374996185, | |
| "learning_rate": 8.321585465494349e-05, | |
| "loss": 0.0158, | |
| "step": 174 | |
| }, | |
| { | |
| "epoch": 1.7656441717791411, | |
| "grad_norm": 0.07977139949798584, | |
| "learning_rate": 8.163601067896343e-05, | |
| "loss": 0.0154, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.8245398773006136, | |
| "grad_norm": 0.07933256775140762, | |
| "learning_rate": 8.000163992119146e-05, | |
| "loss": 0.0203, | |
| "step": 186 | |
| }, | |
| { | |
| "epoch": 1.883435582822086, | |
| "grad_norm": 0.047850318253040314, | |
| "learning_rate": 7.831555932901642e-05, | |
| "loss": 0.0146, | |
| "step": 192 | |
| }, | |
| { | |
| "epoch": 1.9423312883435582, | |
| "grad_norm": 0.07077929377555847, | |
| "learning_rate": 7.658067497518772e-05, | |
| "loss": 0.0164, | |
| "step": 198 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 0.08206511288881302, | |
| "learning_rate": 7.479997704900437e-05, | |
| "loss": 0.0164, | |
| "step": 204 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.022676732391119003, | |
| "eval_runtime": 102.7598, | |
| "eval_samples_per_second": 0.886, | |
| "eval_steps_per_second": 0.886, | |
| "step": 204 | |
| } | |
| ], | |
| "logging_steps": 6, | |
| "max_steps": 505, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "save_steps": 51, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 3.8128925519806464e+17, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |