| { | |
| "best_metric": 1.1237905025482178, | |
| "best_model_checkpoint": "model_llama2_7b_pre_fa_v2/checkpoint-30000", | |
| "epoch": 1.5, | |
| "eval_steps": 6000, | |
| "global_step": 36000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 1.95e-05, | |
| "loss": 1.4763, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "eval_loss": 1.215256690979004, | |
| "eval_runtime": 3808.4515, | |
| "eval_samples_per_second": 1.575, | |
| "eval_steps_per_second": 1.575, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 1.9e-05, | |
| "loss": 1.194, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "eval_loss": 1.1560847759246826, | |
| "eval_runtime": 4331.4845, | |
| "eval_samples_per_second": 1.385, | |
| "eval_steps_per_second": 1.385, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 1.8500000000000002e-05, | |
| "loss": 1.1423, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "eval_loss": 1.1338403224945068, | |
| "eval_runtime": 4323.9918, | |
| "eval_samples_per_second": 1.388, | |
| "eval_steps_per_second": 1.388, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 1.8e-05, | |
| "loss": 1.1314, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 1.1248250007629395, | |
| "eval_runtime": 3781.6164, | |
| "eval_samples_per_second": 1.587, | |
| "eval_steps_per_second": 1.587, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "learning_rate": 1.7500000000000002e-05, | |
| "loss": 1.1134, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "eval_loss": 1.1237905025482178, | |
| "eval_runtime": 4344.7683, | |
| "eval_samples_per_second": 1.381, | |
| "eval_steps_per_second": 1.381, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "learning_rate": 1.7e-05, | |
| "loss": 1.1102, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "eval_loss": 1.1263515949249268, | |
| "eval_runtime": 4343.9602, | |
| "eval_samples_per_second": 1.381, | |
| "eval_steps_per_second": 1.381, | |
| "step": 36000 | |
| } | |
| ], | |
| "logging_steps": 6000, | |
| "max_steps": 240000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 6000, | |
| "total_flos": 2.9590328180736e+18, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |