| { |
| "best_global_step": null, |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 0.002916727431821496, |
| "eval_steps": 500, |
| "global_step": 5, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "clip_ratio/high_max": 0.0, |
| "clip_ratio/high_mean": 0.0, |
| "clip_ratio/low_mean": 0.0, |
| "clip_ratio/low_min": 0.0, |
| "clip_ratio/region_mean": 0.0, |
| "completions/clipped_ratio": 0.825, |
| "completions/max_length": 256.0, |
| "completions/max_terminated_length": 207.0, |
| "completions/mean_length": 253.421875, |
| "completions/mean_terminated_length": 138.1428680419922, |
| "completions/min_length": 1.0, |
| "completions/min_terminated_length": 1.0, |
| "epoch": 0.0005833454863642993, |
| "frac_reward_zero_std": 0.0, |
| "grad_norm": 5.4375, |
| "kl": 0.0, |
| "learning_rate": 0.0, |
| "loss": 0.0006, |
| "num_tokens": 100507.0, |
| "reward": 1.2124054431915283, |
| "reward_std": 0.2885759472846985, |
| "rewards/wrapper/mean": 1.2124054431915283, |
| "rewards/wrapper/std": 0.4714105725288391, |
| "step": 1 |
| }, |
| { |
| "clip_ratio/high_max": 0.0, |
| "clip_ratio/high_mean": 0.0, |
| "clip_ratio/low_mean": 0.0, |
| "clip_ratio/low_min": 0.0, |
| "clip_ratio/region_mean": 0.0, |
| "completions/clipped_ratio": 0.775, |
| "completions/max_length": 256.0, |
| "completions/max_terminated_length": 229.0, |
| "completions/mean_length": 253.24063110351562, |
| "completions/mean_terminated_length": 157.88888549804688, |
| "completions/min_length": 77.0, |
| "completions/min_terminated_length": 77.0, |
| "epoch": 0.0011666909727285986, |
| "frac_reward_zero_std": 0.0062500000931322575, |
| "grad_norm": 6.75, |
| "kl": 0.0, |
| "learning_rate": 2.1428571428571427e-06, |
| "loss": -0.0033, |
| "num_tokens": 198892.0, |
| "reward": 1.269961953163147, |
| "reward_std": 0.26232200860977173, |
| "rewards/wrapper/mean": 1.269961953163147, |
| "rewards/wrapper/std": 0.5202347040176392, |
| "step": 2 |
| }, |
| { |
| "clip_ratio/high_max": 0.0, |
| "clip_ratio/high_mean": 0.0, |
| "clip_ratio/low_mean": 0.0, |
| "clip_ratio/low_min": 0.0, |
| "clip_ratio/region_mean": 0.0, |
| "completions/clipped_ratio": 0.725, |
| "completions/max_length": 256.0, |
| "completions/max_terminated_length": 206.0, |
| "completions/mean_length": 251.38125610351562, |
| "completions/mean_terminated_length": 121.63636779785156, |
| "completions/min_length": 1.0, |
| "completions/min_terminated_length": 1.0, |
| "epoch": 0.0017500364590928978, |
| "frac_reward_zero_std": 0.0062500000931322575, |
| "grad_norm": 6.0, |
| "kl": 0.001783137078746222, |
| "learning_rate": 4.2857142857142855e-06, |
| "loss": -0.0048, |
| "num_tokens": 301180.0, |
| "reward": 1.2353280782699585, |
| "reward_std": 0.2479369193315506, |
| "rewards/wrapper/mean": 1.2353280782699585, |
| "rewards/wrapper/std": 0.4726465046405792, |
| "step": 3 |
| }, |
| { |
| "clip_ratio/high_max": 0.0, |
| "clip_ratio/high_mean": 0.0, |
| "clip_ratio/low_mean": 0.0, |
| "clip_ratio/low_min": 0.0, |
| "clip_ratio/region_mean": 0.0, |
| "completions/clipped_ratio": 0.8, |
| "completions/max_length": 256.0, |
| "completions/max_terminated_length": 227.0, |
| "completions/mean_length": 253.25, |
| "completions/mean_terminated_length": 146.0, |
| "completions/min_length": 34.0, |
| "completions/min_terminated_length": 34.0, |
| "epoch": 0.002333381945457197, |
| "frac_reward_zero_std": 0.0, |
| "grad_norm": 8.5, |
| "kl": 0.02622605988290161, |
| "learning_rate": 6.428571428571428e-06, |
| "loss": 0.0001, |
| "num_tokens": 405512.0, |
| "reward": 1.262830376625061, |
| "reward_std": 0.30002185702323914, |
| "rewards/wrapper/mean": 1.2628302574157715, |
| "rewards/wrapper/std": 0.5151814818382263, |
| "step": 4 |
| }, |
| { |
| "clip_ratio/high_max": 0.0, |
| "clip_ratio/high_mean": 0.0, |
| "clip_ratio/low_mean": 0.0, |
| "clip_ratio/low_min": 0.0, |
| "clip_ratio/region_mean": 0.0, |
| "completions/clipped_ratio": 0.7, |
| "completions/max_length": 256.0, |
| "completions/max_terminated_length": 243.0, |
| "completions/mean_length": 252.2624969482422, |
| "completions/mean_terminated_length": 156.33334350585938, |
| "completions/min_length": 26.0, |
| "completions/min_terminated_length": 26.0, |
| "epoch": 0.002916727431821496, |
| "frac_reward_zero_std": 0.0062500000931322575, |
| "grad_norm": 9.25, |
| "kl": 0.05060198903083801, |
| "learning_rate": 8.571428571428571e-06, |
| "loss": 0.0017, |
| "num_tokens": 505254.0, |
| "reward": 1.227962851524353, |
| "reward_std": 0.2680162191390991, |
| "rewards/wrapper/mean": 1.227962851524353, |
| "rewards/wrapper/std": 0.4670800566673279, |
| "step": 5 |
| }, |
| { |
| "epoch": 0.002916727431821496, |
| "eval_clip_ratio/high_max": 0.0, |
| "eval_clip_ratio/high_mean": 0.0, |
| "eval_clip_ratio/low_mean": 0.0, |
| "eval_clip_ratio/low_min": 0.0, |
| "eval_clip_ratio/region_mean": 0.0, |
| "eval_completions/clipped_ratio": 0.8269230769230769, |
| "eval_completions/max_length": 256.0, |
| "eval_completions/max_terminated_length": 90.76923076923077, |
| "eval_completions/mean_length": 254.00721153846155, |
| "eval_completions/mean_terminated_length": 80.32051321176382, |
| "eval_completions/min_length": 204.69230769230768, |
| "eval_completions/min_terminated_length": 66.84615384615384, |
| "eval_frac_reward_zero_std": 0.0, |
| "eval_kl": 0.06102241260501055, |
| "eval_loss": 0.0018880075076594949, |
| "eval_num_tokens": 505254.0, |
| "eval_reward": 1.274492282133836, |
| "eval_reward_std": 0.27295459233797514, |
| "eval_rewards/wrapper/mean": 1.274492282133836, |
| "eval_rewards/wrapper/std": 0.46636236630953276, |
| "eval_runtime": 19.138, |
| "eval_samples_per_second": 10.45, |
| "eval_steps_per_second": 0.366, |
| "step": 5 |
| } |
| ], |
| "logging_steps": 1.0, |
| "max_steps": 3428, |
| "num_input_tokens_seen": 505254, |
| "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": false |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 0.0, |
| "train_batch_size": 10, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
