| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.026055237102657634, | |
| "eval_steps": 4, | |
| "global_step": 25, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0010422094841063053, | |
| "grad_norm": 3.42587947845459, | |
| "learning_rate": 1.0000000000000002e-06, | |
| "loss": 6.4631, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0010422094841063053, | |
| "eval_loss": 6.671364784240723, | |
| "eval_runtime": 9.44, | |
| "eval_samples_per_second": 42.903, | |
| "eval_steps_per_second": 10.805, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0020844189682126106, | |
| "grad_norm": 3.787745714187622, | |
| "learning_rate": 2.0000000000000003e-06, | |
| "loss": 6.619, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.003126628452318916, | |
| "grad_norm": 3.6949379444122314, | |
| "learning_rate": 3e-06, | |
| "loss": 6.5476, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.004168837936425221, | |
| "grad_norm": 4.343378067016602, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 6.6851, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.004168837936425221, | |
| "eval_loss": 6.672913074493408, | |
| "eval_runtime": 7.8311, | |
| "eval_samples_per_second": 51.717, | |
| "eval_steps_per_second": 13.025, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.005211047420531527, | |
| "grad_norm": 4.449336528778076, | |
| "learning_rate": 5e-06, | |
| "loss": 6.7603, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.006253256904637832, | |
| "grad_norm": 4.981034755706787, | |
| "learning_rate": 6e-06, | |
| "loss": 6.7741, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.007295466388744137, | |
| "grad_norm": 4.62288236618042, | |
| "learning_rate": 7e-06, | |
| "loss": 7.3853, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.008337675872850442, | |
| "grad_norm": 4.797128677368164, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 7.1027, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.008337675872850442, | |
| "eval_loss": 6.669874668121338, | |
| "eval_runtime": 7.7933, | |
| "eval_samples_per_second": 51.968, | |
| "eval_steps_per_second": 13.088, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.009379885356956748, | |
| "grad_norm": 4.196491718292236, | |
| "learning_rate": 9e-06, | |
| "loss": 6.7359, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.010422094841063054, | |
| "grad_norm": 3.4589948654174805, | |
| "learning_rate": 1e-05, | |
| "loss": 6.5552, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.011464304325169358, | |
| "grad_norm": 3.497850179672241, | |
| "learning_rate": 9.890738003669029e-06, | |
| "loss": 6.8426, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.012506513809275664, | |
| "grad_norm": 3.9026501178741455, | |
| "learning_rate": 9.567727288213005e-06, | |
| "loss": 6.9201, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.012506513809275664, | |
| "eval_loss": 6.656294822692871, | |
| "eval_runtime": 7.8499, | |
| "eval_samples_per_second": 51.593, | |
| "eval_steps_per_second": 12.994, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.01354872329338197, | |
| "grad_norm": 3.805992603302002, | |
| "learning_rate": 9.045084971874738e-06, | |
| "loss": 6.4677, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.014590932777488274, | |
| "grad_norm": 3.989011526107788, | |
| "learning_rate": 8.345653031794292e-06, | |
| "loss": 6.8528, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.015633142261594582, | |
| "grad_norm": 4.253493785858154, | |
| "learning_rate": 7.500000000000001e-06, | |
| "loss": 6.8437, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.016675351745700884, | |
| "grad_norm": 4.157368183135986, | |
| "learning_rate": 6.545084971874738e-06, | |
| "loss": 7.0895, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.016675351745700884, | |
| "eval_loss": 6.632657051086426, | |
| "eval_runtime": 7.8431, | |
| "eval_samples_per_second": 51.638, | |
| "eval_steps_per_second": 13.005, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.01771756122980719, | |
| "grad_norm": 4.2801594734191895, | |
| "learning_rate": 5.522642316338268e-06, | |
| "loss": 6.8602, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.018759770713913496, | |
| "grad_norm": 3.379889726638794, | |
| "learning_rate": 4.477357683661734e-06, | |
| "loss": 6.3006, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.019801980198019802, | |
| "grad_norm": 4.244846820831299, | |
| "learning_rate": 3.4549150281252635e-06, | |
| "loss": 6.6127, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.020844189682126108, | |
| "grad_norm": 4.060556411743164, | |
| "learning_rate": 2.5000000000000015e-06, | |
| "loss": 6.3001, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.020844189682126108, | |
| "eval_loss": 6.617997646331787, | |
| "eval_runtime": 7.8463, | |
| "eval_samples_per_second": 51.617, | |
| "eval_steps_per_second": 13.0, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.021886399166232414, | |
| "grad_norm": 4.195801258087158, | |
| "learning_rate": 1.6543469682057105e-06, | |
| "loss": 7.1121, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.022928608650338717, | |
| "grad_norm": 3.2973010540008545, | |
| "learning_rate": 9.549150281252633e-07, | |
| "loss": 5.6943, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.023970818134445022, | |
| "grad_norm": 3.863523244857788, | |
| "learning_rate": 4.322727117869951e-07, | |
| "loss": 6.3595, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.02501302761855133, | |
| "grad_norm": 3.6141159534454346, | |
| "learning_rate": 1.0926199633097156e-07, | |
| "loss": 6.1388, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.02501302761855133, | |
| "eval_loss": 6.613474369049072, | |
| "eval_runtime": 7.81, | |
| "eval_samples_per_second": 51.856, | |
| "eval_steps_per_second": 13.06, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.026055237102657634, | |
| "grad_norm": 4.716870307922363, | |
| "learning_rate": 0.0, | |
| "loss": 6.6627, | |
| "step": 25 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 25, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 25, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 347093965209600.0, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |