| { |
| "best_global_step": null, |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 0.125, |
| "eval_steps": 500, |
| "global_step": 50, |
| "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": 1.0, |
| "completions/max_length": 256.0, |
| "completions/max_terminated_length": 0.0, |
| "completions/mean_length": 256.0, |
| "completions/mean_terminated_length": 0.0, |
| "completions/min_length": 256.0, |
| "completions/min_terminated_length": 0.0, |
| "entropy": 1.975826621055603, |
| "epoch": 0.0025, |
| "frac_reward_zero_std": 1.0, |
| "grad_norm": 0.0, |
| "learning_rate": 1e-05, |
| "loss": 0.0, |
| "num_tokens": 1132.0, |
| "reward": 0.009999999776482582, |
| "reward_std": 0.0, |
| "rewards/autonomy_reward_fn/mean": 0.009999999776482582, |
| "rewards/autonomy_reward_fn/std": 0.0, |
| "step": 1, |
| "step_time": 5.940139313000145 |
| }, |
| { |
| "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, |
| "entropy": 2.439243793487549, |
| "epoch": 0.005, |
| "grad_norm": 0.0, |
| "learning_rate": 9.800000000000001e-06, |
| "loss": 0.0, |
| "step": 2, |
| "step_time": 0.13269777900040935 |
| }, |
| { |
| "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, |
| "entropy": 2.167536497116089, |
| "epoch": 0.0075, |
| "grad_norm": 0.0, |
| "learning_rate": 9.600000000000001e-06, |
| "loss": 0.0, |
| "step": 3, |
| "step_time": 0.12384193100024277 |
| }, |
| { |
| "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, |
| "entropy": 2.270764112472534, |
| "epoch": 0.01, |
| "grad_norm": 0.0, |
| "learning_rate": 9.4e-06, |
| "loss": 0.0, |
| "step": 4, |
| "step_time": 0.12020948200006387 |
| }, |
| { |
| "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": 1.0, |
| "completions/max_length": 256.0, |
| "completions/max_terminated_length": 0.0, |
| "completions/mean_length": 256.0, |
| "completions/mean_terminated_length": 0.0, |
| "completions/min_length": 256.0, |
| "completions/min_terminated_length": 0.0, |
| "entropy": 1.7375574111938477, |
| "epoch": 0.0125, |
| "frac_reward_zero_std": 1.0, |
| "grad_norm": 0.0, |
| "learning_rate": 9.200000000000002e-06, |
| "loss": 0.0, |
| "num_tokens": 2240.0, |
| "reward": 0.009999999776482582, |
| "reward_std": 0.0, |
| "rewards/autonomy_reward_fn/mean": 0.009999999776482582, |
| "rewards/autonomy_reward_fn/std": 0.0, |
| "step": 5, |
| "step_time": 5.4328740460005065 |
| }, |
| { |
| "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, |
| "entropy": 1.2307178974151611, |
| "epoch": 0.015, |
| "grad_norm": 0.0, |
| "learning_rate": 9e-06, |
| "loss": 0.0, |
| "step": 6, |
| "step_time": 0.12100409000049694 |
| }, |
| { |
| "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, |
| "entropy": 2.5965301990509033, |
| "epoch": 0.0175, |
| "grad_norm": 0.0, |
| "learning_rate": 8.8e-06, |
| "loss": 0.0, |
| "step": 7, |
| "step_time": 0.11992437099979725 |
| }, |
| { |
| "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, |
| "entropy": 3.000967502593994, |
| "epoch": 0.02, |
| "grad_norm": 0.0, |
| "learning_rate": 8.6e-06, |
| "loss": 0.0, |
| "step": 8, |
| "step_time": 0.12026366800000687 |
| }, |
| { |
| "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": 1.0, |
| "completions/max_length": 256.0, |
| "completions/max_terminated_length": 0.0, |
| "completions/mean_length": 256.0, |
| "completions/mean_terminated_length": 0.0, |
| "completions/min_length": 256.0, |
| "completions/min_terminated_length": 0.0, |
| "entropy": 2.1747303009033203, |
| "epoch": 0.0225, |
| "frac_reward_zero_std": 1.0, |
| "grad_norm": 0.0, |
| "learning_rate": 8.400000000000001e-06, |
| "loss": 0.0, |
| "num_tokens": 3372.0, |
| "reward": 0.009999999776482582, |
| "reward_std": 0.0, |
| "rewards/autonomy_reward_fn/mean": 0.009999999776482582, |
| "rewards/autonomy_reward_fn/std": 0.0, |
| "step": 9, |
| "step_time": 5.488657728999897 |
| }, |
| { |
| "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, |
| "entropy": 2.2987818717956543, |
| "epoch": 0.025, |
| "grad_norm": 0.0, |
| "learning_rate": 8.2e-06, |
| "loss": 0.0, |
| "step": 10, |
| "step_time": 0.12148796400015272 |
| }, |
| { |
| "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, |
| "entropy": 2.549589157104492, |
| "epoch": 0.0275, |
| "grad_norm": 0.0, |
| "learning_rate": 8.000000000000001e-06, |
| "loss": 0.0, |
| "step": 11, |
| "step_time": 0.1202768709999873 |
| }, |
| { |
| "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, |
| "entropy": 2.3748786449432373, |
| "epoch": 0.03, |
| "grad_norm": 0.0, |
| "learning_rate": 7.800000000000002e-06, |
| "loss": 0.0, |
| "step": 12, |
| "step_time": 0.12146498200036149 |
| }, |
| { |
| "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": 1.0, |
| "completions/max_length": 256.0, |
| "completions/max_terminated_length": 0.0, |
| "completions/mean_length": 256.0, |
| "completions/mean_terminated_length": 0.0, |
| "completions/min_length": 256.0, |
| "completions/min_terminated_length": 0.0, |
| "entropy": 2.562595844268799, |
| "epoch": 0.0325, |
| "frac_reward_zero_std": 1.0, |
| "grad_norm": 0.0, |
| "learning_rate": 7.600000000000001e-06, |
| "loss": 0.0, |
| "num_tokens": 4504.0, |
| "reward": 0.009999999776482582, |
| "reward_std": 0.0, |
| "rewards/autonomy_reward_fn/mean": 0.009999999776482582, |
| "rewards/autonomy_reward_fn/std": 0.0, |
| "step": 13, |
| "step_time": 5.529226956000457 |
| }, |
| { |
| "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, |
| "entropy": 2.9105567932128906, |
| "epoch": 0.035, |
| "grad_norm": 0.0, |
| "learning_rate": 7.4e-06, |
| "loss": 0.0, |
| "step": 14, |
| "step_time": 0.12003547699987394 |
| }, |
| { |
| "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, |
| "entropy": 2.360198497772217, |
| "epoch": 0.0375, |
| "grad_norm": 0.0, |
| "learning_rate": 7.2000000000000005e-06, |
| "loss": 0.0, |
| "step": 15, |
| "step_time": 0.12056965900046634 |
| }, |
| { |
| "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, |
| "entropy": 1.9906266927719116, |
| "epoch": 0.04, |
| "grad_norm": 0.0, |
| "learning_rate": 7e-06, |
| "loss": 0.0, |
| "step": 16, |
| "step_time": 0.11991975299952173 |
| }, |
| { |
| "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": 1.0, |
| "completions/max_length": 256.0, |
| "completions/max_terminated_length": 0.0, |
| "completions/mean_length": 256.0, |
| "completions/mean_terminated_length": 0.0, |
| "completions/min_length": 256.0, |
| "completions/min_terminated_length": 0.0, |
| "entropy": 2.4847490787506104, |
| "epoch": 0.0425, |
| "frac_reward_zero_std": 1.0, |
| "grad_norm": 0.0, |
| "learning_rate": 6.800000000000001e-06, |
| "loss": 0.0, |
| "num_tokens": 5636.0, |
| "reward": 0.009999999776482582, |
| "reward_std": 0.0, |
| "rewards/autonomy_reward_fn/mean": 0.009999999776482582, |
| "rewards/autonomy_reward_fn/std": 0.0, |
| "step": 17, |
| "step_time": 5.571549678000338 |
| }, |
| { |
| "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, |
| "entropy": 2.0457265377044678, |
| "epoch": 0.045, |
| "grad_norm": 0.0, |
| "learning_rate": 6.600000000000001e-06, |
| "loss": 0.0, |
| "step": 18, |
| "step_time": 0.12112131499998213 |
| }, |
| { |
| "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, |
| "entropy": 2.4426045417785645, |
| "epoch": 0.0475, |
| "grad_norm": 0.0, |
| "learning_rate": 6.4000000000000006e-06, |
| "loss": 0.0, |
| "step": 19, |
| "step_time": 0.12062360999971133 |
| }, |
| { |
| "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, |
| "entropy": 2.8799967765808105, |
| "epoch": 0.05, |
| "grad_norm": 0.0, |
| "learning_rate": 6.200000000000001e-06, |
| "loss": 0.0, |
| "step": 20, |
| "step_time": 0.1196468329999334 |
| }, |
| { |
| "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": 1.0, |
| "completions/max_length": 256.0, |
| "completions/max_terminated_length": 0.0, |
| "completions/mean_length": 256.0, |
| "completions/mean_terminated_length": 0.0, |
| "completions/min_length": 256.0, |
| "completions/min_terminated_length": 0.0, |
| "entropy": 2.4007647037506104, |
| "epoch": 0.0525, |
| "frac_reward_zero_std": 1.0, |
| "grad_norm": 0.0, |
| "learning_rate": 6e-06, |
| "loss": 0.0, |
| "num_tokens": 6756.0, |
| "reward": 0.009999999776482582, |
| "reward_std": 0.0, |
| "rewards/autonomy_reward_fn/mean": 0.009999999776482582, |
| "rewards/autonomy_reward_fn/std": 0.0, |
| "step": 21, |
| "step_time": 5.46478837199993 |
| }, |
| { |
| "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, |
| "entropy": 2.0130887031555176, |
| "epoch": 0.055, |
| "grad_norm": 0.0, |
| "learning_rate": 5.8e-06, |
| "loss": 0.0, |
| "step": 22, |
| "step_time": 0.12104618500052311 |
| }, |
| { |
| "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, |
| "entropy": 3.6849136352539062, |
| "epoch": 0.0575, |
| "grad_norm": 0.0, |
| "learning_rate": 5.600000000000001e-06, |
| "loss": 0.0, |
| "step": 23, |
| "step_time": 0.12017933200058906 |
| }, |
| { |
| "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, |
| "entropy": 2.6699423789978027, |
| "epoch": 0.06, |
| "grad_norm": 0.0, |
| "learning_rate": 5.400000000000001e-06, |
| "loss": 0.0, |
| "step": 24, |
| "step_time": 0.11987412399957975 |
| }, |
| { |
| "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": 1.0, |
| "completions/max_length": 256.0, |
| "completions/max_terminated_length": 0.0, |
| "completions/mean_length": 256.0, |
| "completions/mean_terminated_length": 0.0, |
| "completions/min_length": 256.0, |
| "completions/min_terminated_length": 0.0, |
| "entropy": 2.353423595428467, |
| "epoch": 0.0625, |
| "frac_reward_zero_std": 1.0, |
| "grad_norm": 0.0, |
| "learning_rate": 5.2e-06, |
| "loss": 0.0, |
| "num_tokens": 7864.0, |
| "reward": 0.009999999776482582, |
| "reward_std": 0.0, |
| "rewards/autonomy_reward_fn/mean": 0.009999999776482582, |
| "rewards/autonomy_reward_fn/std": 0.0, |
| "step": 25, |
| "step_time": 5.475499750999916 |
| }, |
| { |
| "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, |
| "entropy": 1.6273622512817383, |
| "epoch": 0.065, |
| "grad_norm": 0.0, |
| "learning_rate": 5e-06, |
| "loss": 0.0, |
| "step": 26, |
| "step_time": 0.12144228599936469 |
| }, |
| { |
| "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, |
| "entropy": 1.7559497356414795, |
| "epoch": 0.0675, |
| "grad_norm": 0.0, |
| "learning_rate": 4.800000000000001e-06, |
| "loss": 0.0, |
| "step": 27, |
| "step_time": 0.12019816300016828 |
| }, |
| { |
| "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, |
| "entropy": 3.4082813262939453, |
| "epoch": 0.07, |
| "grad_norm": 0.0, |
| "learning_rate": 4.600000000000001e-06, |
| "loss": 0.0, |
| "step": 28, |
| "step_time": 0.12150602700057789 |
| }, |
| { |
| "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": 1.0, |
| "completions/max_length": 256.0, |
| "completions/max_terminated_length": 0.0, |
| "completions/mean_length": 256.0, |
| "completions/mean_terminated_length": 0.0, |
| "completions/min_length": 256.0, |
| "completions/min_terminated_length": 0.0, |
| "entropy": 2.0362539291381836, |
| "epoch": 0.0725, |
| "frac_reward_zero_std": 1.0, |
| "grad_norm": 0.0, |
| "learning_rate": 4.4e-06, |
| "loss": 0.0, |
| "num_tokens": 8972.0, |
| "reward": 0.009999999776482582, |
| "reward_std": 0.0, |
| "rewards/autonomy_reward_fn/mean": 0.009999999776482582, |
| "rewards/autonomy_reward_fn/std": 0.0, |
| "step": 29, |
| "step_time": 5.578138597999896 |
| }, |
| { |
| "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, |
| "entropy": 1.9473274946212769, |
| "epoch": 0.075, |
| "grad_norm": 0.0, |
| "learning_rate": 4.2000000000000004e-06, |
| "loss": 0.0, |
| "step": 30, |
| "step_time": 0.11901681000017561 |
| }, |
| { |
| "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, |
| "entropy": 3.764465570449829, |
| "epoch": 0.0775, |
| "grad_norm": 0.0, |
| "learning_rate": 4.000000000000001e-06, |
| "loss": 0.0, |
| "step": 31, |
| "step_time": 0.11898399900019285 |
| }, |
| { |
| "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, |
| "entropy": 1.9196826219558716, |
| "epoch": 0.08, |
| "grad_norm": 0.0, |
| "learning_rate": 3.8000000000000005e-06, |
| "loss": 0.0, |
| "step": 32, |
| "step_time": 0.12208616299994901 |
| }, |
| { |
| "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": 1.0, |
| "completions/max_length": 256.0, |
| "completions/max_terminated_length": 0.0, |
| "completions/mean_length": 256.0, |
| "completions/mean_terminated_length": 0.0, |
| "completions/min_length": 256.0, |
| "completions/min_terminated_length": 0.0, |
| "entropy": 2.1861605644226074, |
| "epoch": 0.0825, |
| "frac_reward_zero_std": 1.0, |
| "grad_norm": 0.0, |
| "learning_rate": 3.6000000000000003e-06, |
| "loss": 0.0, |
| "num_tokens": 10104.0, |
| "reward": 0.009999999776482582, |
| "reward_std": 0.0, |
| "rewards/autonomy_reward_fn/mean": 0.009999999776482582, |
| "rewards/autonomy_reward_fn/std": 0.0, |
| "step": 33, |
| "step_time": 5.526427269999658 |
| }, |
| { |
| "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, |
| "entropy": 1.9254075288772583, |
| "epoch": 0.085, |
| "grad_norm": 0.0, |
| "learning_rate": 3.4000000000000005e-06, |
| "loss": 0.0, |
| "step": 34, |
| "step_time": 0.11869688299975678 |
| }, |
| { |
| "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, |
| "entropy": 2.2586331367492676, |
| "epoch": 0.0875, |
| "grad_norm": 0.0, |
| "learning_rate": 3.2000000000000003e-06, |
| "loss": 0.0, |
| "step": 35, |
| "step_time": 0.11914809700010665 |
| }, |
| { |
| "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, |
| "entropy": 2.4637088775634766, |
| "epoch": 0.09, |
| "grad_norm": 0.0, |
| "learning_rate": 3e-06, |
| "loss": 0.0, |
| "step": 36, |
| "step_time": 0.1263795000004393 |
| }, |
| { |
| "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": 1.0, |
| "completions/max_length": 256.0, |
| "completions/max_terminated_length": 0.0, |
| "completions/mean_length": 256.0, |
| "completions/mean_terminated_length": 0.0, |
| "completions/min_length": 256.0, |
| "completions/min_terminated_length": 0.0, |
| "entropy": 3.3892905712127686, |
| "epoch": 0.0925, |
| "frac_reward_zero_std": 1.0, |
| "grad_norm": 0.0, |
| "learning_rate": 2.8000000000000003e-06, |
| "loss": 0.0, |
| "num_tokens": 11212.0, |
| "reward": 0.009999999776482582, |
| "reward_std": 0.0, |
| "rewards/autonomy_reward_fn/mean": 0.009999999776482582, |
| "rewards/autonomy_reward_fn/std": 0.0, |
| "step": 37, |
| "step_time": 5.478073480000603 |
| }, |
| { |
| "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, |
| "entropy": 3.1603095531463623, |
| "epoch": 0.095, |
| "grad_norm": 0.0, |
| "learning_rate": 2.6e-06, |
| "loss": 0.0, |
| "step": 38, |
| "step_time": 0.12008928099930927 |
| }, |
| { |
| "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, |
| "entropy": 2.8118975162506104, |
| "epoch": 0.0975, |
| "grad_norm": 0.0, |
| "learning_rate": 2.4000000000000003e-06, |
| "loss": 0.0, |
| "step": 39, |
| "step_time": 0.11953222700049082 |
| }, |
| { |
| "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, |
| "entropy": 1.9064228534698486, |
| "epoch": 0.1, |
| "grad_norm": 0.0, |
| "learning_rate": 2.2e-06, |
| "loss": 0.0, |
| "step": 40, |
| "step_time": 0.11825968199991621 |
| }, |
| { |
| "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": 1.0, |
| "completions/max_length": 256.0, |
| "completions/max_terminated_length": 0.0, |
| "completions/mean_length": 256.0, |
| "completions/mean_terminated_length": 0.0, |
| "completions/min_length": 256.0, |
| "completions/min_terminated_length": 0.0, |
| "entropy": 2.5093579292297363, |
| "epoch": 0.1025, |
| "frac_reward_zero_std": 1.0, |
| "grad_norm": 0.0, |
| "learning_rate": 2.0000000000000003e-06, |
| "loss": 0.0, |
| "num_tokens": 12320.0, |
| "reward": 0.009999999776482582, |
| "reward_std": 0.0, |
| "rewards/autonomy_reward_fn/mean": 0.009999999776482582, |
| "rewards/autonomy_reward_fn/std": 0.0, |
| "step": 41, |
| "step_time": 5.509690339999906 |
| }, |
| { |
| "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, |
| "entropy": 2.3179931640625, |
| "epoch": 0.105, |
| "grad_norm": 0.0, |
| "learning_rate": 1.8000000000000001e-06, |
| "loss": 0.0, |
| "step": 42, |
| "step_time": 0.12237605300015275 |
| }, |
| { |
| "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, |
| "entropy": 1.4812008142471313, |
| "epoch": 0.1075, |
| "grad_norm": 0.0, |
| "learning_rate": 1.6000000000000001e-06, |
| "loss": 0.0, |
| "step": 43, |
| "step_time": 0.12001185999997688 |
| }, |
| { |
| "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, |
| "entropy": 4.90458869934082, |
| "epoch": 0.11, |
| "grad_norm": 0.0, |
| "learning_rate": 1.4000000000000001e-06, |
| "loss": 0.0, |
| "step": 44, |
| "step_time": 0.11975958299990452 |
| }, |
| { |
| "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": 1.0, |
| "completions/max_length": 256.0, |
| "completions/max_terminated_length": 0.0, |
| "completions/mean_length": 256.0, |
| "completions/mean_terminated_length": 0.0, |
| "completions/min_length": 256.0, |
| "completions/min_terminated_length": 0.0, |
| "entropy": 2.2333452701568604, |
| "epoch": 0.1125, |
| "frac_reward_zero_std": 1.0, |
| "grad_norm": 0.0, |
| "learning_rate": 1.2000000000000002e-06, |
| "loss": 0.0, |
| "num_tokens": 13452.0, |
| "reward": 0.009999999776482582, |
| "reward_std": 0.0, |
| "rewards/autonomy_reward_fn/mean": 0.009999999776482582, |
| "rewards/autonomy_reward_fn/std": 0.0, |
| "step": 45, |
| "step_time": 5.512979769000594 |
| }, |
| { |
| "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, |
| "entropy": 1.8680591583251953, |
| "epoch": 0.115, |
| "grad_norm": 0.0, |
| "learning_rate": 1.0000000000000002e-06, |
| "loss": 0.0, |
| "step": 46, |
| "step_time": 0.12362773699987883 |
| }, |
| { |
| "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, |
| "entropy": 3.232543706893921, |
| "epoch": 0.1175, |
| "grad_norm": 0.0, |
| "learning_rate": 8.000000000000001e-07, |
| "loss": 0.0, |
| "step": 47, |
| "step_time": 0.12175974799993128 |
| }, |
| { |
| "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, |
| "entropy": 2.694847583770752, |
| "epoch": 0.12, |
| "grad_norm": 0.0, |
| "learning_rate": 6.000000000000001e-07, |
| "loss": 0.0, |
| "step": 48, |
| "step_time": 0.11931696000010561 |
| }, |
| { |
| "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": 1.0, |
| "completions/max_length": 256.0, |
| "completions/max_terminated_length": 0.0, |
| "completions/mean_length": 256.0, |
| "completions/mean_terminated_length": 0.0, |
| "completions/min_length": 256.0, |
| "completions/min_terminated_length": 0.0, |
| "entropy": 2.182255268096924, |
| "epoch": 0.1225, |
| "frac_reward_zero_std": 1.0, |
| "grad_norm": 0.0, |
| "learning_rate": 4.0000000000000003e-07, |
| "loss": 0.0, |
| "num_tokens": 14572.0, |
| "reward": 0.009999999776482582, |
| "reward_std": 0.0, |
| "rewards/autonomy_reward_fn/mean": 0.009999999776482582, |
| "rewards/autonomy_reward_fn/std": 0.0, |
| "step": 49, |
| "step_time": 5.534827849000067 |
| }, |
| { |
| "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, |
| "entropy": 2.535165309906006, |
| "epoch": 0.125, |
| "grad_norm": 0.0, |
| "learning_rate": 2.0000000000000002e-07, |
| "loss": 0.0, |
| "step": 50, |
| "step_time": 0.12092450500040286 |
| } |
| ], |
| "logging_steps": 1, |
| "max_steps": 50, |
| "num_input_tokens_seen": 14572, |
| "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": 0.0, |
| "train_batch_size": 1, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
