| { | |
| "best_global_step": null, | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.1, | |
| "eval_steps": 500, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "completion_length": 256.0, | |
| "epoch": 0.01, | |
| "grad_norm": 0.40769532322883606, | |
| "kl": 1.5046377666294574e-05, | |
| "learning_rate": 4.5000000000000003e-07, | |
| "loss": 0.0, | |
| "reward": -2.465625, | |
| "reward_std": 0.42179486304521563, | |
| "rewards/reward_check_answer": 0.0, | |
| "rewards/reward_check_numbers": 0.0, | |
| "rewards/reward_consciousness": 0.284375, | |
| "rewards/reward_format_approximately": -2.75, | |
| "rewards/reward_format_exactly": 0.0, | |
| "rewards/reward_inner_working": 0.0, | |
| "step": 10 | |
| }, | |
| { | |
| "completion_length": 256.0, | |
| "epoch": 0.02, | |
| "grad_norm": 0.4385190010070801, | |
| "kl": 1.570033491589129e-05, | |
| "learning_rate": 9.500000000000001e-07, | |
| "loss": 0.0, | |
| "reward": -2.31875, | |
| "reward_std": 0.5976869776844979, | |
| "rewards/reward_check_answer": 0.0, | |
| "rewards/reward_check_numbers": 0.0, | |
| "rewards/reward_consciousness": 0.30625, | |
| "rewards/reward_format_approximately": -2.625, | |
| "rewards/reward_format_exactly": 0.0, | |
| "rewards/reward_inner_working": 0.0, | |
| "step": 20 | |
| }, | |
| { | |
| "completion_length": 256.0, | |
| "epoch": 0.03, | |
| "grad_norm": 0.4469616115093231, | |
| "kl": 1.6215385403484107e-05, | |
| "learning_rate": 1.45e-06, | |
| "loss": 0.0, | |
| "reward": -2.459375, | |
| "reward_std": 0.3866558074951172, | |
| "rewards/reward_check_answer": 0.0, | |
| "rewards/reward_check_numbers": 0.0, | |
| "rewards/reward_consciousness": 0.315625, | |
| "rewards/reward_format_approximately": -2.775, | |
| "rewards/reward_format_exactly": 0.0, | |
| "rewards/reward_inner_working": 0.0, | |
| "step": 30 | |
| }, | |
| { | |
| "completion_length": 256.0, | |
| "epoch": 0.04, | |
| "grad_norm": 0.4186990559101105, | |
| "kl": 1.6400672029703857e-05, | |
| "learning_rate": 1.9500000000000004e-06, | |
| "loss": 0.0, | |
| "reward": -2.221875, | |
| "reward_std": 0.7143462687730789, | |
| "rewards/reward_check_answer": 0.0, | |
| "rewards/reward_check_numbers": 0.0, | |
| "rewards/reward_consciousness": 0.278125, | |
| "rewards/reward_format_approximately": -2.5, | |
| "rewards/reward_format_exactly": 0.0, | |
| "rewards/reward_inner_working": 0.0, | |
| "step": 40 | |
| }, | |
| { | |
| "completion_length": 256.0, | |
| "epoch": 0.05, | |
| "grad_norm": 0.44865837693214417, | |
| "kl": 1.9841926405206323e-05, | |
| "learning_rate": 2.4500000000000003e-06, | |
| "loss": 0.0, | |
| "reward": -2.39375, | |
| "reward_std": 0.5951654136180877, | |
| "rewards/reward_check_answer": 0.0, | |
| "rewards/reward_check_numbers": 0.0, | |
| "rewards/reward_consciousness": 0.25625, | |
| "rewards/reward_format_approximately": -2.65, | |
| "rewards/reward_format_exactly": 0.0, | |
| "rewards/reward_inner_working": 0.0, | |
| "step": 50 | |
| }, | |
| { | |
| "completion_length": 255.3, | |
| "epoch": 0.06, | |
| "grad_norm": 0.4461980164051056, | |
| "kl": 2.5847879624052438e-05, | |
| "learning_rate": 2.95e-06, | |
| "loss": 0.0, | |
| "reward": -2.240625, | |
| "reward_std": 0.7081772074103355, | |
| "rewards/reward_check_answer": 0.0, | |
| "rewards/reward_check_numbers": 0.0, | |
| "rewards/reward_consciousness": 0.309375, | |
| "rewards/reward_format_approximately": -2.55, | |
| "rewards/reward_format_exactly": 0.0, | |
| "rewards/reward_inner_working": 0.0, | |
| "step": 60 | |
| }, | |
| { | |
| "completion_length": 256.0, | |
| "epoch": 0.07, | |
| "grad_norm": 0.3938167691230774, | |
| "kl": 3.849279601126909e-05, | |
| "learning_rate": 3.45e-06, | |
| "loss": 0.0, | |
| "reward": -2.384375, | |
| "reward_std": 0.3940433248877525, | |
| "rewards/reward_check_answer": 0.0, | |
| "rewards/reward_check_numbers": 0.0, | |
| "rewards/reward_consciousness": 0.340625, | |
| "rewards/reward_format_approximately": -2.725, | |
| "rewards/reward_format_exactly": 0.0, | |
| "rewards/reward_inner_working": 0.0, | |
| "step": 70 | |
| }, | |
| { | |
| "completion_length": 256.0, | |
| "epoch": 0.08, | |
| "grad_norm": 0.44823428988456726, | |
| "kl": 6.826544413343071e-05, | |
| "learning_rate": 3.95e-06, | |
| "loss": 0.0, | |
| "reward": -2.296875, | |
| "reward_std": 0.5537523284554482, | |
| "rewards/reward_check_answer": 0.0, | |
| "rewards/reward_check_numbers": 0.0, | |
| "rewards/reward_consciousness": 0.353125, | |
| "rewards/reward_format_approximately": -2.65, | |
| "rewards/reward_format_exactly": 0.0, | |
| "rewards/reward_inner_working": 0.0, | |
| "step": 80 | |
| }, | |
| { | |
| "completion_length": 256.0, | |
| "epoch": 0.09, | |
| "grad_norm": 0.4049831032752991, | |
| "kl": 9.327681036666035e-05, | |
| "learning_rate": 4.450000000000001e-06, | |
| "loss": 0.0, | |
| "reward": -2.471875, | |
| "reward_std": 0.33873592466115954, | |
| "rewards/reward_check_answer": 0.0, | |
| "rewards/reward_check_numbers": 0.0, | |
| "rewards/reward_consciousness": 0.328125, | |
| "rewards/reward_format_approximately": -2.8, | |
| "rewards/reward_format_exactly": 0.0, | |
| "rewards/reward_inner_working": 0.0, | |
| "step": 90 | |
| }, | |
| { | |
| "completion_length": 256.0, | |
| "epoch": 0.1, | |
| "grad_norm": 0.43748927116394043, | |
| "kl": 0.00017035281634889544, | |
| "learning_rate": 4.95e-06, | |
| "loss": 0.0, | |
| "reward": -2.26875, | |
| "reward_std": 0.5575396627187729, | |
| "rewards/reward_check_answer": 0.0, | |
| "rewards/reward_check_numbers": 0.0, | |
| "rewards/reward_consciousness": 0.28125, | |
| "rewards/reward_format_approximately": -2.55, | |
| "rewards/reward_format_exactly": 0.0, | |
| "rewards/reward_inner_working": 0.0, | |
| "step": 100 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 1000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 100, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 0.0, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |