| { |
| "best_global_step": null, |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 0.4, |
| "eval_steps": 500, |
| "global_step": 100, |
| "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": 0.08333333333333337, |
| "completions/max_length": 391.0, |
| "completions/max_terminated_length": 391.0, |
| "completions/mean_length": 202.91666666666666, |
| "completions/mean_terminated_length": 221.36363636363637, |
| "completions/min_length": 0.0, |
| "completions/min_terminated_length": 11.0, |
| "epoch": 0.004, |
| "format_failures": 0.0, |
| "grad_norm": 0.0, |
| "kl": 0.0, |
| "learning_rate": 0.0, |
| "loss": 0.0, |
| "num_tokens": 18672.0, |
| "reward": 0.0, |
| "reward_std": 0.0, |
| "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": 0.08333333333333337, |
| "completions/max_length": 161.0, |
| "completions/max_terminated_length": 161.0, |
| "completions/mean_length": 92.83333333333333, |
| "completions/mean_terminated_length": 101.27272727272727, |
| "completions/min_length": 0.0, |
| "completions/min_terminated_length": 13.0, |
| "epoch": 0.008, |
| "format_failures": 0.0, |
| "grad_norm": 0.0, |
| "kl": 0.0, |
| "learning_rate": 1e-06, |
| "loss": 0.0, |
| "num_tokens": 29988.0, |
| "reward": 0.0, |
| "reward_std": 0.0, |
| "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": 0.08333333333333337, |
| "completions/max_length": 151.0, |
| "completions/max_terminated_length": 151.0, |
| "completions/mean_length": 52.333333333333336, |
| "completions/mean_terminated_length": 57.09090909090909, |
| "completions/min_length": 0.0, |
| "completions/min_terminated_length": 13.0, |
| "epoch": 0.012, |
| "format_failures": 0.0, |
| "grad_norm": 0.0, |
| "kl": 0.0, |
| "learning_rate": 1e-06, |
| "loss": 0.0, |
| "num_tokens": 39576.0, |
| "reward": 0.0, |
| "reward_std": 0.0, |
| "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": 0.08333333333333337, |
| "completions/max_length": 326.0, |
| "completions/max_terminated_length": 326.0, |
| "completions/mean_length": 161.75, |
| "completions/mean_terminated_length": 176.45454545454547, |
| "completions/min_length": 0.0, |
| "completions/min_terminated_length": 18.0, |
| "epoch": 0.016, |
| "format_failures": 0.0, |
| "grad_norm": 0.0, |
| "kl": 0.0, |
| "learning_rate": 1e-06, |
| "loss": 0.0, |
| "num_tokens": 53340.0, |
| "reward": 0.0, |
| "reward_std": 0.0, |
| "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": 0.08333333333333337, |
| "completions/max_length": 120.0, |
| "completions/max_terminated_length": 120.0, |
| "completions/mean_length": 75.08333333333333, |
| "completions/mean_terminated_length": 81.9090909090909, |
| "completions/min_length": 0.0, |
| "completions/min_terminated_length": 11.0, |
| "epoch": 0.02, |
| "format_failures": 0.0, |
| "grad_norm": 0.0, |
| "kl": 0.0, |
| "learning_rate": 1e-06, |
| "loss": 0.0, |
| "num_tokens": 60420.0, |
| "reward": 0.0, |
| "reward_std": 0.0, |
| "step": 5 |
| }, |
| { |
| "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": 0.08333333333333337, |
| "completions/max_length": 374.0, |
| "completions/max_terminated_length": 374.0, |
| "completions/mean_length": 178.0, |
| "completions/mean_terminated_length": 194.1818181818182, |
| "completions/min_length": 0.0, |
| "completions/min_terminated_length": 11.0, |
| "epoch": 0.024, |
| "format_failures": 0.0, |
| "grad_norm": 0.0, |
| "kl": 0.0, |
| "learning_rate": 1e-06, |
| "loss": 0.0, |
| "num_tokens": 75348.0, |
| "reward": 0.0, |
| "reward_std": 0.0, |
| "step": 6 |
| }, |
| { |
| "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": 0.08333333333333337, |
| "completions/max_length": 146.0, |
| "completions/max_terminated_length": 146.0, |
| "completions/mean_length": 99.08333333333333, |
| "completions/mean_terminated_length": 108.0909090909091, |
| "completions/min_length": 0.0, |
| "completions/min_terminated_length": 94.0, |
| "epoch": 0.028, |
| "format_failures": 0.0, |
| "grad_norm": 1.6362388134002686, |
| "kl": 0.0, |
| "learning_rate": 1e-06, |
| "loss": 0.0806, |
| "num_tokens": 83868.0, |
| "reward": 0.5833333730697632, |
| "reward_std": 0.5149286389350891, |
| "step": 7 |
| }, |
| { |
| "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": 0.08333333333333337, |
| "completions/max_length": 358.0, |
| "completions/max_terminated_length": 358.0, |
| "completions/mean_length": 190.0, |
| "completions/mean_terminated_length": 207.27272727272728, |
| "completions/min_length": 0.0, |
| "completions/min_terminated_length": 124.0, |
| "epoch": 0.032, |
| "format_failures": 0.0, |
| "grad_norm": 0.0022762538865208626, |
| "kl": 0.0005378490750445053, |
| "learning_rate": 1e-06, |
| "loss": 0.0, |
| "num_tokens": 97464.0, |
| "reward": 0.0, |
| "reward_std": 0.0, |
| "step": 8 |
| }, |
| { |
| "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": 0.08333333333333337, |
| "completions/max_length": 245.0, |
| "completions/max_terminated_length": 245.0, |
| "completions/mean_length": 141.16666666666666, |
| "completions/mean_terminated_length": 154.0, |
| "completions/min_length": 0.0, |
| "completions/min_terminated_length": 13.0, |
| "epoch": 0.036, |
| "format_failures": 0.0, |
| "grad_norm": 0.007215190213173628, |
| "kl": 0.0019240143010392785, |
| "learning_rate": 1e-06, |
| "loss": 0.0, |
| "num_tokens": 108636.0, |
| "reward": 0.0, |
| "reward_std": 0.0, |
| "step": 9 |
| }, |
| { |
| "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": 0.08333333333333337, |
| "completions/max_length": 266.0, |
| "completions/max_terminated_length": 266.0, |
| "completions/mean_length": 178.41666666666666, |
| "completions/mean_terminated_length": 194.63636363636363, |
| "completions/min_length": 0.0, |
| "completions/min_terminated_length": 142.0, |
| "epoch": 0.04, |
| "format_failures": 0.0, |
| "grad_norm": 0.7695807218551636, |
| "kl": 0.014113324228674173, |
| "learning_rate": 1e-06, |
| "loss": -0.002, |
| "num_tokens": 116256.0, |
| "reward": 0.1666666716337204, |
| "reward_std": 0.38924944400787354, |
| "step": 10 |
| }, |
| { |
| "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": 0.08333333333333337, |
| "completions/max_length": 135.0, |
| "completions/max_terminated_length": 135.0, |
| "completions/mean_length": 84.25, |
| "completions/mean_terminated_length": 91.9090909090909, |
| "completions/min_length": 0.0, |
| "completions/min_terminated_length": 10.0, |
| "epoch": 0.044, |
| "format_failures": 0.0, |
| "grad_norm": 1.026847243309021, |
| "kl": 0.013075211551040411, |
| "learning_rate": 1e-06, |
| "loss": -0.0413, |
| "num_tokens": 124440.0, |
| "reward": 0.1666666716337204, |
| "reward_std": 0.38924944400787354, |
| "step": 11 |
| }, |
| { |
| "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": 0.08333333333333337, |
| "completions/max_length": 199.0, |
| "completions/max_terminated_length": 199.0, |
| "completions/mean_length": 99.25, |
| "completions/mean_terminated_length": 108.27272727272727, |
| "completions/min_length": 0.0, |
| "completions/min_terminated_length": 10.0, |
| "epoch": 0.048, |
| "format_failures": 1.0, |
| "grad_norm": 0.011818243190646172, |
| "kl": 0.003624255710747093, |
| "learning_rate": 1e-06, |
| "loss": 0.0, |
| "num_tokens": 132732.0, |
| "reward": 0.0, |
| "reward_std": 0.0, |
| "step": 12 |
| }, |
| { |
| "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": 0.08333333333333337, |
| "completions/max_length": 577.0, |
| "completions/max_terminated_length": 577.0, |
| "completions/mean_length": 246.08333333333334, |
| "completions/mean_terminated_length": 268.45454545454544, |
| "completions/min_length": 0.0, |
| "completions/min_terminated_length": 10.0, |
| "epoch": 0.052, |
| "format_failures": 0.0, |
| "grad_norm": 0.022241737693548203, |
| "kl": 0.00960063119418919, |
| "learning_rate": 1e-06, |
| "loss": 0.0, |
| "num_tokens": 152424.0, |
| "reward": 0.0, |
| "reward_std": 0.0, |
| "step": 13 |
| }, |
| { |
| "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": 0.08333333333333337, |
| "completions/max_length": 344.0, |
| "completions/max_terminated_length": 344.0, |
| "completions/mean_length": 173.0, |
| "completions/mean_terminated_length": 188.72727272727272, |
| "completions/min_length": 0.0, |
| "completions/min_terminated_length": 117.0, |
| "epoch": 0.056, |
| "format_failures": 0.0, |
| "grad_norm": 0.036054644733667374, |
| "kl": 0.01761903613805771, |
| "learning_rate": 1e-06, |
| "loss": 0.0001, |
| "num_tokens": 162636.0, |
| "reward": 0.0, |
| "reward_std": 0.0, |
| "step": 14 |
| }, |
| { |
| "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": 0.08333333333333337, |
| "completions/max_length": 402.0, |
| "completions/max_terminated_length": 402.0, |
| "completions/mean_length": 268.1666666666667, |
| "completions/mean_terminated_length": 292.54545454545456, |
| "completions/min_length": 0.0, |
| "completions/min_terminated_length": 102.0, |
| "epoch": 0.06, |
| "format_failures": 0.0, |
| "grad_norm": 0.00860360637307167, |
| "kl": 0.005409976467490196, |
| "learning_rate": 1e-06, |
| "loss": 0.0, |
| "num_tokens": 176904.0, |
| "reward": 0.0, |
| "reward_std": 0.0, |
| "step": 15 |
| }, |
| { |
| "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": 0.08333333333333337, |
| "completions/max_length": 127.0, |
| "completions/max_terminated_length": 127.0, |
| "completions/mean_length": 72.16666666666667, |
| "completions/mean_terminated_length": 78.72727272727273, |
| "completions/min_length": 0.0, |
| "completions/min_terminated_length": 10.0, |
| "epoch": 0.064, |
| "format_failures": 0.0, |
| "grad_norm": 0.8470466136932373, |
| "kl": 0.09470756724476814, |
| "learning_rate": 1e-06, |
| "loss": 0.0173, |
| "num_tokens": 186564.0, |
| "reward": 0.0833333358168602, |
| "reward_std": 0.28867512941360474, |
| "step": 16 |
| }, |
| { |
| "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": 0.08333333333333337, |
| "completions/max_length": 245.0, |
| "completions/max_terminated_length": 245.0, |
| "completions/mean_length": 118.5, |
| "completions/mean_terminated_length": 129.27272727272728, |
| "completions/min_length": 0.0, |
| "completions/min_terminated_length": 34.0, |
| "epoch": 0.068, |
| "format_failures": 0.0, |
| "grad_norm": 0.10086339712142944, |
| "kl": 0.04859759844839573, |
| "learning_rate": 1e-06, |
| "loss": 0.0003, |
| "num_tokens": 197484.0, |
| "reward": 0.0, |
| "reward_std": 0.0, |
| "step": 17 |
| }, |
| { |
| "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": 0.08333333333333337, |
| "completions/max_length": 88.0, |
| "completions/max_terminated_length": 88.0, |
| "completions/mean_length": 57.833333333333336, |
| "completions/mean_terminated_length": 63.09090909090909, |
| "completions/min_length": 0.0, |
| "completions/min_terminated_length": 12.0, |
| "epoch": 0.072, |
| "format_failures": 0.0, |
| "grad_norm": 1.4592796564102173, |
| "kl": 0.010172125417739153, |
| "learning_rate": 1e-06, |
| "loss": 0.0087, |
| "num_tokens": 207252.0, |
| "reward": 0.0833333358168602, |
| "reward_std": 0.28867512941360474, |
| "step": 18 |
| }, |
| { |
| "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": 0.08333333333333337, |
| "completions/max_length": 57.0, |
| "completions/max_terminated_length": 57.0, |
| "completions/mean_length": 32.583333333333336, |
| "completions/mean_terminated_length": 35.54545454545455, |
| "completions/min_length": 0.0, |
| "completions/min_terminated_length": 10.0, |
| "epoch": 0.076, |
| "format_failures": 0.0, |
| "grad_norm": 2.4069900512695312, |
| "kl": 0.025834742933511734, |
| "learning_rate": 1e-06, |
| "loss": -0.0695, |
| "num_tokens": 214320.0, |
| "reward": 0.6666666865348816, |
| "reward_std": 0.4923659861087799, |
| "step": 19 |
| }, |
| { |
| "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": 0.08333333333333337, |
| "completions/max_length": 386.0, |
| "completions/max_terminated_length": 386.0, |
| "completions/mean_length": 192.41666666666666, |
| "completions/mean_terminated_length": 209.9090909090909, |
| "completions/min_length": 0.0, |
| "completions/min_terminated_length": 32.0, |
| "epoch": 0.08, |
| "format_failures": 0.0, |
| "grad_norm": 0.10245665162801743, |
| "kl": 0.043199990526773036, |
| "learning_rate": 1e-06, |
| "loss": 0.0002, |
| "num_tokens": 228996.0, |
| "reward": 0.0, |
| "reward_std": 0.0, |
| "step": 20 |
| }, |
| { |
| "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": 0.16666666666666663, |
| "completions/max_length": 88.0, |
| "completions/max_terminated_length": 88.0, |
| "completions/mean_length": 67.91666666666667, |
| "completions/mean_terminated_length": 81.5, |
| "completions/min_length": 0.0, |
| "completions/min_terminated_length": 74.0, |
| "epoch": 0.084, |
| "format_failures": 1.0, |
| "grad_norm": 1.388899326324463, |
| "kl": 0.07192051783204079, |
| "learning_rate": 1e-06, |
| "loss": -0.0112, |
| "num_tokens": 238104.0, |
| "reward": 0.75, |
| "reward_std": 0.45226702094078064, |
| "step": 21 |
| }, |
| { |
| "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": 0.08333333333333337, |
| "completions/max_length": 73.0, |
| "completions/max_terminated_length": 73.0, |
| "completions/mean_length": 43.583333333333336, |
| "completions/mean_terminated_length": 47.54545454545455, |
| "completions/min_length": 0.0, |
| "completions/min_terminated_length": 10.0, |
| "epoch": 0.088, |
| "format_failures": 0.0, |
| "grad_norm": 3.2448337078094482, |
| "kl": 0.0771165993064642, |
| "learning_rate": 1e-06, |
| "loss": 0.0077, |
| "num_tokens": 245280.0, |
| "reward": 0.5, |
| "reward_std": 0.5222329497337341, |
| "step": 22 |
| }, |
| { |
| "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": 0.08333333333333337, |
| "completions/max_length": 335.0, |
| "completions/max_terminated_length": 335.0, |
| "completions/mean_length": 167.83333333333334, |
| "completions/mean_terminated_length": 183.0909090909091, |
| "completions/min_length": 0.0, |
| "completions/min_terminated_length": 13.0, |
| "epoch": 0.092, |
| "format_failures": 0.0, |
| "grad_norm": 1.0195705890655518, |
| "kl": 0.211347796022892, |
| "learning_rate": 1e-06, |
| "loss": 0.0009, |
| "num_tokens": 257148.0, |
| "reward": 0.0, |
| "reward_std": 0.0, |
| "step": 23 |
| }, |
| { |
| "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": 0.08333333333333337, |
| "completions/max_length": 350.0, |
| "completions/max_terminated_length": 350.0, |
| "completions/mean_length": 201.41666666666666, |
| "completions/mean_terminated_length": 219.72727272727272, |
| "completions/min_length": 0.0, |
| "completions/min_terminated_length": 101.0, |
| "epoch": 0.096, |
| "format_failures": 0.0, |
| "grad_norm": 0.20492610335350037, |
| "kl": 0.08658944815397263, |
| "learning_rate": 1e-06, |
| "loss": 0.0003, |
| "num_tokens": 266304.0, |
| "reward": 0.0, |
| "reward_std": 0.0, |
| "step": 24 |
| }, |
| { |
| "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": 0.08333333333333337, |
| "completions/max_length": 312.0, |
| "completions/max_terminated_length": 312.0, |
| "completions/mean_length": 170.0, |
| "completions/mean_terminated_length": 185.45454545454547, |
| "completions/min_length": 0.0, |
| "completions/min_terminated_length": 10.0, |
| "epoch": 0.1, |
| "format_failures": 0.0, |
| "grad_norm": 0.0755978599190712, |
| "kl": 0.040397679433226585, |
| "learning_rate": 1e-06, |
| "loss": 0.0002, |
| "num_tokens": 278760.0, |
| "reward": 0.0, |
| "reward_std": 0.0, |
| "step": 25 |
| }, |
| { |
| "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": 0.08333333333333337, |
| "completions/max_length": 411.0, |
| "completions/max_terminated_length": 411.0, |
| "completions/mean_length": 173.83333333333334, |
| "completions/mean_terminated_length": 189.63636363636363, |
| "completions/min_length": 0.0, |
| "completions/min_terminated_length": 10.0, |
| "epoch": 0.104, |
| "format_failures": 0.0, |
| "grad_norm": 0.04659981280565262, |
| "kl": 0.023209942039102316, |
| "learning_rate": 1e-06, |
| "loss": 0.0001, |
| "num_tokens": 293628.0, |
| "reward": 0.0, |
| "reward_std": 0.0, |
| "step": 26 |
| }, |
| { |
| "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": 0.08333333333333337, |
| "completions/max_length": 92.0, |
| "completions/max_terminated_length": 92.0, |
| "completions/mean_length": 55.25, |
| "completions/mean_terminated_length": 60.27272727272727, |
| "completions/min_length": 0.0, |
| "completions/min_terminated_length": 43.0, |
| "epoch": 0.108, |
| "format_failures": 0.0, |
| "grad_norm": 21.968534469604492, |
| "kl": 5.299874305725098, |
| "learning_rate": 1e-06, |
| "loss": 0.1192, |
| "num_tokens": 301488.0, |
| "reward": 0.5833333730697632, |
| "reward_std": 0.5149286389350891, |
| "step": 27 |
| }, |
| { |
| "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": 0.08333333333333337, |
| "completions/max_length": 358.0, |
| "completions/max_terminated_length": 358.0, |
| "completions/mean_length": 165.41666666666666, |
| "completions/mean_terminated_length": 180.45454545454547, |
| "completions/min_length": 0.0, |
| "completions/min_terminated_length": 11.0, |
| "epoch": 0.112, |
| "format_failures": 0.0, |
| "grad_norm": 0.014507513493299484, |
| "kl": 0.01523882569745183, |
| "learning_rate": 1e-06, |
| "loss": 0.0001, |
| "num_tokens": 314748.0, |
| "reward": 0.0, |
| "reward_std": 0.0, |
| "step": 28 |
| }, |
| { |
| "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": 0.08333333333333337, |
| "completions/max_length": 289.0, |
| "completions/max_terminated_length": 289.0, |
| "completions/mean_length": 193.33333333333334, |
| "completions/mean_terminated_length": 210.9090909090909, |
| "completions/min_length": 0.0, |
| "completions/min_terminated_length": 119.0, |
| "epoch": 0.116, |
| "format_failures": 0.0, |
| "grad_norm": 0.010872351005673409, |
| "kl": 0.010655859019607306, |
| "learning_rate": 1e-06, |
| "loss": 0.0001, |
| "num_tokens": 328692.0, |
| "reward": 0.0, |
| "reward_std": 0.0, |
| "step": 29 |
| }, |
| { |
| "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": 0.08333333333333337, |
| "completions/max_length": 271.0, |
| "completions/max_terminated_length": 271.0, |
| "completions/mean_length": 167.16666666666666, |
| "completions/mean_terminated_length": 182.36363636363637, |
| "completions/min_length": 0.0, |
| "completions/min_terminated_length": 8.0, |
| "epoch": 0.12, |
| "format_failures": 0.0, |
| "grad_norm": 1.0025266408920288, |
| "kl": 0.025600655004382133, |
| "learning_rate": 1e-06, |
| "loss": -0.0472, |
| "num_tokens": 340752.0, |
| "reward": 0.3333333432674408, |
| "reward_std": 0.4923659861087799, |
| "step": 30 |
| }, |
| { |
| "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": 0.08333333333333337, |
| "completions/max_length": 162.0, |
| "completions/max_terminated_length": 162.0, |
| "completions/mean_length": 79.33333333333333, |
| "completions/mean_terminated_length": 86.54545454545455, |
| "completions/min_length": 0.0, |
| "completions/min_terminated_length": 17.0, |
| "epoch": 0.124, |
| "format_failures": 0.0, |
| "grad_norm": 0.01500674244016409, |
| "kl": 0.006932976422831416, |
| "learning_rate": 1e-06, |
| "loss": 0.0, |
| "num_tokens": 366936.0, |
| "reward": 0.0, |
| "reward_std": 0.0, |
| "step": 31 |
| }, |
| { |
| "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": 0.08333333333333337, |
| "completions/max_length": 153.0, |
| "completions/max_terminated_length": 153.0, |
| "completions/mean_length": 100.25, |
| "completions/mean_terminated_length": 109.36363636363636, |
| "completions/min_length": 0.0, |
| "completions/min_terminated_length": 13.0, |
| "epoch": 0.128, |
| "format_failures": 0.0, |
| "grad_norm": 0.572136640548706, |
| "kl": 0.016836593858897686, |
| "learning_rate": 1e-06, |
| "loss": -0.0253, |
| "num_tokens": 375948.0, |
| "reward": 0.0833333358168602, |
| "reward_std": 0.28867512941360474, |
| "step": 32 |
| }, |
| { |
| "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": 0.08333333333333337, |
| "completions/max_length": 91.0, |
| "completions/max_terminated_length": 91.0, |
| "completions/mean_length": 52.75, |
| "completions/mean_terminated_length": 57.54545454545455, |
| "completions/min_length": 0.0, |
| "completions/min_terminated_length": 11.0, |
| "epoch": 0.132, |
| "format_failures": 0.0, |
| "grad_norm": 2.6049137115478516, |
| "kl": 0.08474422618746758, |
| "learning_rate": 1e-06, |
| "loss": -0.0534, |
| "num_tokens": 382608.0, |
| "reward": 0.3333333432674408, |
| "reward_std": 0.4923659861087799, |
| "step": 33 |
| }, |
| { |
| "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": 0.08333333333333337, |
| "completions/max_length": 98.0, |
| "completions/max_terminated_length": 98.0, |
| "completions/mean_length": 61.416666666666664, |
| "completions/mean_terminated_length": 67.0, |
| "completions/min_length": 0.0, |
| "completions/min_terminated_length": 13.0, |
| "epoch": 0.136, |
| "format_failures": 0.0, |
| "grad_norm": 1.9431159496307373, |
| "kl": 0.04839755780994892, |
| "learning_rate": 1e-06, |
| "loss": -0.1095, |
| "num_tokens": 389208.0, |
| "reward": 0.1666666716337204, |
| "reward_std": 0.38924944400787354, |
| "step": 34 |
| }, |
| { |
| "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": 0.08333333333333337, |
| "completions/max_length": 151.0, |
| "completions/max_terminated_length": 151.0, |
| "completions/mean_length": 118.91666666666667, |
| "completions/mean_terminated_length": 129.72727272727272, |
| "completions/min_length": 0.0, |
| "completions/min_terminated_length": 105.0, |
| "epoch": 0.14, |
| "format_failures": 0.0, |
| "grad_norm": 0.03593799099326134, |
| "kl": 0.03462314326316118, |
| "learning_rate": 1e-06, |
| "loss": 0.0002, |
| "num_tokens": 396696.0, |
| "reward": 0.0, |
| "reward_std": 0.0, |
| "step": 35 |
| }, |
| { |
| "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": 0.08333333333333337, |
| "completions/max_length": 371.0, |
| "completions/max_terminated_length": 371.0, |
| "completions/mean_length": 245.58333333333334, |
| "completions/mean_terminated_length": 267.90909090909093, |
| "completions/min_length": 0.0, |
| "completions/min_terminated_length": 183.0, |
| "epoch": 0.144, |
| "format_failures": 0.0, |
| "grad_norm": 0.025885488837957382, |
| "kl": 0.02637413516640663, |
| "learning_rate": 1e-06, |
| "loss": 0.0001, |
| "num_tokens": 411372.0, |
| "reward": 0.0, |
| "reward_std": 0.0, |
| "step": 36 |
| }, |
| { |
| "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": 0.08333333333333337, |
| "completions/max_length": 296.0, |
| "completions/max_terminated_length": 296.0, |
| "completions/mean_length": 179.41666666666666, |
| "completions/mean_terminated_length": 195.72727272727272, |
| "completions/min_length": 0.0, |
| "completions/min_terminated_length": 129.0, |
| "epoch": 0.148, |
| "format_failures": 0.0, |
| "grad_norm": 0.11734314262866974, |
| "kl": 0.0526489345356822, |
| "learning_rate": 1e-06, |
| "loss": 0.0003, |
| "num_tokens": 424404.0, |
| "reward": 0.0, |
| "reward_std": 0.0, |
| "step": 37 |
| }, |
| { |
| "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": 0.08333333333333337, |
| "completions/max_length": 630.0, |
| "completions/max_terminated_length": 630.0, |
| "completions/mean_length": 332.0833333333333, |
| "completions/mean_terminated_length": 362.27272727272725, |
| "completions/min_length": 0.0, |
| "completions/min_terminated_length": 11.0, |
| "epoch": 0.152, |
| "format_failures": 1.0, |
| "grad_norm": 0.5079672932624817, |
| "kl": 0.052276700269430876, |
| "learning_rate": 1e-06, |
| "loss": 0.0148, |
| "num_tokens": 444576.0, |
| "reward": 0.0833333358168602, |
| "reward_std": 0.28867512941360474, |
| "step": 38 |
| }, |
| { |
| "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": 0.08333333333333337, |
| "completions/max_length": 148.0, |
| "completions/max_terminated_length": 148.0, |
| "completions/mean_length": 67.25, |
| "completions/mean_terminated_length": 73.36363636363636, |
| "completions/min_length": 0.0, |
| "completions/min_terminated_length": 8.0, |
| "epoch": 0.156, |
| "format_failures": 0.0, |
| "grad_norm": 2.6541359424591064, |
| "kl": 0.5338308056816459, |
| "learning_rate": 1e-06, |
| "loss": -0.1217, |
| "num_tokens": 453192.0, |
| "reward": 0.25, |
| "reward_std": 0.45226702094078064, |
| "step": 39 |
| }, |
| { |
| "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": 0.08333333333333337, |
| "completions/max_length": 245.0, |
| "completions/max_terminated_length": 245.0, |
| "completions/mean_length": 139.91666666666666, |
| "completions/mean_terminated_length": 152.63636363636363, |
| "completions/min_length": 0.0, |
| "completions/min_terminated_length": 10.0, |
| "epoch": 0.16, |
| "format_failures": 0.0, |
| "grad_norm": 0.3757868707180023, |
| "kl": 0.13857688568532467, |
| "learning_rate": 1e-06, |
| "loss": 0.0009, |
| "num_tokens": 467928.0, |
| "reward": 0.0, |
| "reward_std": 0.0, |
| "step": 40 |
| }, |
| { |
| "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": 0.08333333333333337, |
| "completions/max_length": 84.0, |
| "completions/max_terminated_length": 84.0, |
| "completions/mean_length": 48.0, |
| "completions/mean_terminated_length": 52.36363636363637, |
| "completions/min_length": 0.0, |
| "completions/min_terminated_length": 10.0, |
| "epoch": 0.164, |
| "format_failures": 1.0, |
| "grad_norm": 4.323275566101074, |
| "kl": 0.21433213353157043, |
| "learning_rate": 1e-06, |
| "loss": -0.0993, |
| "num_tokens": 473472.0, |
| "reward": 0.25, |
| "reward_std": 0.45226702094078064, |
| "step": 41 |
| }, |
| { |
| "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": 0.08333333333333337, |
| "completions/max_length": 311.0, |
| "completions/max_terminated_length": 311.0, |
| "completions/mean_length": 109.58333333333333, |
| "completions/mean_terminated_length": 119.54545454545455, |
| "completions/min_length": 0.0, |
| "completions/min_terminated_length": 10.0, |
| "epoch": 0.168, |
| "format_failures": 0.0, |
| "grad_norm": 0.22781899571418762, |
| "kl": 0.07318945415318012, |
| "learning_rate": 1e-06, |
| "loss": 0.0006, |
| "num_tokens": 488148.0, |
| "reward": 0.0, |
| "reward_std": 0.0, |
| "step": 42 |
| }, |
| { |
| "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": 0.08333333333333337, |
| "completions/max_length": 49.0, |
| "completions/max_terminated_length": 49.0, |
| "completions/mean_length": 31.166666666666668, |
| "completions/mean_terminated_length": 34.0, |
| "completions/min_length": 0.0, |
| "completions/min_terminated_length": 10.0, |
| "epoch": 0.172, |
| "format_failures": 0.0, |
| "grad_norm": 2.492840051651001, |
| "kl": 0.224076546728611, |
| "learning_rate": 1e-06, |
| "loss": -0.0283, |
| "num_tokens": 492624.0, |
| "reward": 0.25, |
| "reward_std": 0.45226702094078064, |
| "step": 43 |
| }, |
| { |
| "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": 0.08333333333333337, |
| "completions/max_length": 66.0, |
| "completions/max_terminated_length": 66.0, |
| "completions/mean_length": 52.25, |
| "completions/mean_terminated_length": 57.0, |
| "completions/min_length": 0.0, |
| "completions/min_terminated_length": 32.0, |
| "epoch": 0.176, |
| "format_failures": 0.0, |
| "grad_norm": 3.037781000137329, |
| "kl": 0.2150058075785637, |
| "learning_rate": 1e-06, |
| "loss": -0.013, |
| "num_tokens": 499752.0, |
| "reward": 0.6666666865348816, |
| "reward_std": 0.4923659861087799, |
| "step": 44 |
| }, |
| { |
| "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": 0.08333333333333337, |
| "completions/max_length": 144.0, |
| "completions/max_terminated_length": 144.0, |
| "completions/mean_length": 83.0, |
| "completions/mean_terminated_length": 90.54545454545455, |
| "completions/min_length": 0.0, |
| "completions/min_terminated_length": 11.0, |
| "epoch": 0.18, |
| "format_failures": 0.0, |
| "grad_norm": 2.3224222660064697, |
| "kl": 0.36255764216184616, |
| "learning_rate": 1e-06, |
| "loss": -0.0749, |
| "num_tokens": 508428.0, |
| "reward": 0.0833333358168602, |
| "reward_std": 0.28867512941360474, |
| "step": 45 |
| }, |
| { |
| "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": 0.08333333333333337, |
| "completions/max_length": 160.0, |
| "completions/max_terminated_length": 160.0, |
| "completions/mean_length": 119.0, |
| "completions/mean_terminated_length": 129.8181818181818, |
| "completions/min_length": 0.0, |
| "completions/min_terminated_length": 106.0, |
| "epoch": 0.184, |
| "format_failures": 0.0, |
| "grad_norm": 1.5437301397323608, |
| "kl": 0.06716796010732651, |
| "learning_rate": 1e-06, |
| "loss": 0.0526, |
| "num_tokens": 517416.0, |
| "reward": 0.75, |
| "reward_std": 0.45226702094078064, |
| "step": 46 |
| }, |
| { |
| "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": 0.08333333333333337, |
| "completions/max_length": 162.0, |
| "completions/max_terminated_length": 162.0, |
| "completions/mean_length": 86.66666666666667, |
| "completions/mean_terminated_length": 94.54545454545455, |
| "completions/min_length": 0.0, |
| "completions/min_terminated_length": 11.0, |
| "epoch": 0.188, |
| "format_failures": 0.0, |
| "grad_norm": 1.92403244972229, |
| "kl": 0.04993921332061291, |
| "learning_rate": 1e-06, |
| "loss": -0.0667, |
| "num_tokens": 525384.0, |
| "reward": 0.25, |
| "reward_std": 0.45226702094078064, |
| "step": 47 |
| }, |
| { |
| "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": 0.08333333333333337, |
| "completions/max_length": 607.0, |
| "completions/max_terminated_length": 607.0, |
| "completions/mean_length": 263.0833333333333, |
| "completions/mean_terminated_length": 287.0, |
| "completions/min_length": 0.0, |
| "completions/min_terminated_length": 10.0, |
| "epoch": 0.192, |
| "format_failures": 0.0, |
| "grad_norm": 0.002583070658147335, |
| "kl": 0.0069114591460675, |
| "learning_rate": 1e-06, |
| "loss": 0.0, |
| "num_tokens": 560328.0, |
| "reward": 0.0, |
| "reward_std": 0.0, |
| "step": 48 |
| }, |
| { |
| "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": 0.08333333333333337, |
| "completions/max_length": 337.0, |
| "completions/max_terminated_length": 337.0, |
| "completions/mean_length": 178.16666666666666, |
| "completions/mean_terminated_length": 194.36363636363637, |
| "completions/min_length": 0.0, |
| "completions/min_terminated_length": 9.0, |
| "epoch": 0.196, |
| "format_failures": 0.0, |
| "grad_norm": 0.055018555372953415, |
| "kl": 0.04814303293824196, |
| "learning_rate": 1e-06, |
| "loss": 0.0002, |
| "num_tokens": 573552.0, |
| "reward": 0.0, |
| "reward_std": 0.0, |
| "step": 49 |
| }, |
| { |
| "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": 0.08333333333333337, |
| "completions/max_length": 205.0, |
| "completions/max_terminated_length": 205.0, |
| "completions/mean_length": 104.25, |
| "completions/mean_terminated_length": 113.72727272727273, |
| "completions/min_length": 0.0, |
| "completions/min_terminated_length": 12.0, |
| "epoch": 0.2, |
| "format_failures": 0.0, |
| "grad_norm": 0.10304596275091171, |
| "kl": 0.0782565288245678, |
| "learning_rate": 1e-06, |
| "loss": 0.0005, |
| "num_tokens": 583980.0, |
| "reward": 0.0, |
| "reward_std": 0.0, |
| "step": 50 |
| }, |
| { |
| "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": 0.08333333333333337, |
| "completions/max_length": 529.0, |
| "completions/max_terminated_length": 529.0, |
| "completions/mean_length": 171.25, |
| "completions/mean_terminated_length": 186.8181818181818, |
| "completions/min_length": 0.0, |
| "completions/min_terminated_length": 8.0, |
| "epoch": 0.204, |
| "format_failures": 0.0, |
| "grad_norm": 0.10456845909357071, |
| "kl": 0.05266672745347023, |
| "learning_rate": 1e-06, |
| "loss": 0.0004, |
| "num_tokens": 606264.0, |
| "reward": 0.0, |
| "reward_std": 0.0, |
| "step": 51 |
| }, |
| { |
| "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": 0.08333333333333337, |
| "completions/max_length": 142.0, |
| "completions/max_terminated_length": 142.0, |
| "completions/mean_length": 86.0, |
| "completions/mean_terminated_length": 93.81818181818181, |
| "completions/min_length": 0.0, |
| "completions/min_terminated_length": 11.0, |
| "epoch": 0.208, |
| "format_failures": 0.0, |
| "grad_norm": 2.117820978164673, |
| "kl": 0.12709446623921394, |
| "learning_rate": 1e-06, |
| "loss": -0.0977, |
| "num_tokens": 616176.0, |
| "reward": 0.25, |
| "reward_std": 0.45226702094078064, |
| "step": 52 |
| }, |
| { |
| "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": 0.08333333333333337, |
| "completions/max_length": 169.0, |
| "completions/max_terminated_length": 169.0, |
| "completions/mean_length": 67.66666666666667, |
| "completions/mean_terminated_length": 73.81818181818181, |
| "completions/min_length": 0.0, |
| "completions/min_terminated_length": 26.0, |
| "epoch": 0.212, |
| "format_failures": 0.0, |
| "grad_norm": 0.36178988218307495, |
| "kl": 0.06635316368192434, |
| "learning_rate": 1e-06, |
| "loss": 0.0007, |
| "num_tokens": 625992.0, |
| "reward": 0.0, |
| "reward_std": 0.0, |
| "step": 53 |
| }, |
| { |
| "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": 0.08333333333333337, |
| "completions/max_length": 292.0, |
| "completions/max_terminated_length": 292.0, |
| "completions/mean_length": 180.5, |
| "completions/mean_terminated_length": 196.9090909090909, |
| "completions/min_length": 0.0, |
| "completions/min_terminated_length": 88.0, |
| "epoch": 0.216, |
| "format_failures": 0.0, |
| "grad_norm": 5.520895957946777, |
| "kl": 0.6420021317899227, |
| "learning_rate": 1e-06, |
| "loss": 0.0043, |
| "num_tokens": 640824.0, |
| "reward": 0.0, |
| "reward_std": 0.0, |
| "step": 54 |
| }, |
| { |
| "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": 0.08333333333333337, |
| "completions/max_length": 81.0, |
| "completions/max_terminated_length": 81.0, |
| "completions/mean_length": 59.5, |
| "completions/mean_terminated_length": 64.9090909090909, |
| "completions/min_length": 0.0, |
| "completions/min_terminated_length": 13.0, |
| "epoch": 0.22, |
| "format_failures": 0.0, |
| "grad_norm": 9.40858268737793, |
| "kl": 1.514443002641201, |
| "learning_rate": 1e-06, |
| "loss": -0.0487, |
| "num_tokens": 649008.0, |
| "reward": 0.5833333730697632, |
| "reward_std": 0.5149286389350891, |
| "step": 55 |
| }, |
| { |
| "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": 0.08333333333333337, |
| "completions/max_length": 647.0, |
| "completions/max_terminated_length": 647.0, |
| "completions/mean_length": 265.25, |
| "completions/mean_terminated_length": 289.3636363636364, |
| "completions/min_length": 0.0, |
| "completions/min_terminated_length": 13.0, |
| "epoch": 0.224, |
| "format_failures": 0.0, |
| "grad_norm": 0.12246920168399811, |
| "kl": 0.04888852685689926, |
| "learning_rate": 1e-06, |
| "loss": 0.0002, |
| "num_tokens": 665112.0, |
| "reward": 0.0, |
| "reward_std": 0.0, |
| "step": 56 |
| }, |
| { |
| "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": 0.16666666666666663, |
| "completions/max_length": 90.0, |
| "completions/max_terminated_length": 90.0, |
| "completions/mean_length": 54.416666666666664, |
| "completions/mean_terminated_length": 65.3, |
| "completions/min_length": 0.0, |
| "completions/min_terminated_length": 10.0, |
| "epoch": 0.228, |
| "format_failures": 0.0, |
| "grad_norm": 0.4643149971961975, |
| "kl": 0.2062125913798809, |
| "learning_rate": 1e-06, |
| "loss": 0.0018, |
| "num_tokens": 671268.0, |
| "reward": 0.0, |
| "reward_std": 0.0, |
| "step": 57 |
| }, |
| { |
| "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": 0.08333333333333337, |
| "completions/max_length": 285.0, |
| "completions/max_terminated_length": 285.0, |
| "completions/mean_length": 138.83333333333334, |
| "completions/mean_terminated_length": 151.45454545454547, |
| "completions/min_length": 0.0, |
| "completions/min_terminated_length": 11.0, |
| "epoch": 0.232, |
| "format_failures": 0.0, |
| "grad_norm": 0.028489232063293457, |
| "kl": 0.028692953288555145, |
| "learning_rate": 1e-06, |
| "loss": 0.0001, |
| "num_tokens": 681648.0, |
| "reward": 0.0, |
| "reward_std": 0.0, |
| "step": 58 |
| }, |
| { |
| "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": 0.08333333333333337, |
| "completions/max_length": 189.0, |
| "completions/max_terminated_length": 189.0, |
| "completions/mean_length": 119.33333333333333, |
| "completions/mean_terminated_length": 130.1818181818182, |
| "completions/min_length": 0.0, |
| "completions/min_terminated_length": 13.0, |
| "epoch": 0.236, |
| "format_failures": 1.0, |
| "grad_norm": 0.2943709194660187, |
| "kl": 0.021217118948698044, |
| "learning_rate": 1e-06, |
| "loss": 0.0001, |
| "num_tokens": 692148.0, |
| "reward": 0.0, |
| "reward_std": 0.0, |
| "step": 59 |
| }, |
| { |
| "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": 0.08333333333333337, |
| "completions/max_length": 123.0, |
| "completions/max_terminated_length": 123.0, |
| "completions/mean_length": 65.25, |
| "completions/mean_terminated_length": 71.18181818181819, |
| "completions/min_length": 0.0, |
| "completions/min_terminated_length": 12.0, |
| "epoch": 0.24, |
| "format_failures": 1.0, |
| "grad_norm": 0.4704815149307251, |
| "kl": 0.1355944722890854, |
| "learning_rate": 1e-06, |
| "loss": 0.0016, |
| "num_tokens": 705504.0, |
| "reward": 0.0, |
| "reward_std": 0.0, |
| "step": 60 |
| }, |
| { |
| "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": 0.08333333333333337, |
| "completions/max_length": 209.0, |
| "completions/max_terminated_length": 209.0, |
| "completions/mean_length": 106.75, |
| "completions/mean_terminated_length": 116.45454545454545, |
| "completions/min_length": 0.0, |
| "completions/min_terminated_length": 61.0, |
| "epoch": 0.244, |
| "format_failures": 0.0, |
| "grad_norm": 0.8021370768547058, |
| "kl": 0.06047418341040611, |
| "learning_rate": 1e-06, |
| "loss": 0.0229, |
| "num_tokens": 712920.0, |
| "reward": 0.0833333358168602, |
| "reward_std": 0.28867512941360474, |
| "step": 61 |
| }, |
| { |
| "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": 0.08333333333333337, |
| "completions/max_length": 356.0, |
| "completions/max_terminated_length": 356.0, |
| "completions/mean_length": 201.83333333333334, |
| "completions/mean_terminated_length": 220.1818181818182, |
| "completions/min_length": 0.0, |
| "completions/min_terminated_length": 10.0, |
| "epoch": 0.248, |
| "format_failures": 0.0, |
| "grad_norm": 0.0354565754532814, |
| "kl": 0.051246967166662216, |
| "learning_rate": 1e-06, |
| "loss": 0.0002, |
| "num_tokens": 725280.0, |
| "reward": 0.0, |
| "reward_std": 0.0, |
| "step": 62 |
| }, |
| { |
| "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": 0.08333333333333337, |
| "completions/max_length": 260.0, |
| "completions/max_terminated_length": 260.0, |
| "completions/mean_length": 129.0, |
| "completions/mean_terminated_length": 140.72727272727272, |
| "completions/min_length": 0.0, |
| "completions/min_terminated_length": 11.0, |
| "epoch": 0.252, |
| "format_failures": 0.0, |
| "grad_norm": 0.838399350643158, |
| "kl": 0.03389432094991207, |
| "learning_rate": 1e-06, |
| "loss": 0.0256, |
| "num_tokens": 733980.0, |
| "reward": 0.0833333358168602, |
| "reward_std": 0.28867512941360474, |
| "step": 63 |
| }, |
| { |
| "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": 0.08333333333333337, |
| "completions/max_length": 227.0, |
| "completions/max_terminated_length": 227.0, |
| "completions/mean_length": 121.08333333333333, |
| "completions/mean_terminated_length": 132.0909090909091, |
| "completions/min_length": 0.0, |
| "completions/min_terminated_length": 11.0, |
| "epoch": 0.256, |
| "format_failures": 0.0, |
| "grad_norm": 0.008542679250240326, |
| "kl": 0.02384038269519806, |
| "learning_rate": 1e-06, |
| "loss": 0.0, |
| "num_tokens": 763224.0, |
| "reward": 0.0, |
| "reward_std": 0.0, |
| "step": 64 |
| }, |
| { |
| "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": 0.08333333333333337, |
| "completions/max_length": 302.0, |
| "completions/max_terminated_length": 302.0, |
| "completions/mean_length": 180.33333333333334, |
| "completions/mean_terminated_length": 196.72727272727272, |
| "completions/min_length": 0.0, |
| "completions/min_terminated_length": 36.0, |
| "epoch": 0.26, |
| "format_failures": 0.0, |
| "grad_norm": 0.01127433218061924, |
| "kl": 0.013883833773434162, |
| "learning_rate": 1e-06, |
| "loss": 0.0001, |
| "num_tokens": 777984.0, |
| "reward": 0.0, |
| "reward_std": 0.0, |
| "step": 65 |
| }, |
| { |
| "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": 0.08333333333333337, |
| "completions/max_length": 141.0, |
| "completions/max_terminated_length": 141.0, |
| "completions/mean_length": 75.83333333333333, |
| "completions/mean_terminated_length": 82.72727272727273, |
| "completions/min_length": 0.0, |
| "completions/min_terminated_length": 21.0, |
| "epoch": 0.264, |
| "format_failures": 0.0, |
| "grad_norm": 0.09972423315048218, |
| "kl": 0.06396586634218693, |
| "learning_rate": 1e-06, |
| "loss": 0.0005, |
| "num_tokens": 785844.0, |
| "reward": 0.0, |
| "reward_std": 0.0, |
| "step": 66 |
| }, |
| { |
| "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": 0.08333333333333337, |
| "completions/max_length": 222.0, |
| "completions/max_terminated_length": 222.0, |
| "completions/mean_length": 140.41666666666666, |
| "completions/mean_terminated_length": 153.1818181818182, |
| "completions/min_length": 0.0, |
| "completions/min_terminated_length": 13.0, |
| "epoch": 0.268, |
| "format_failures": 0.0, |
| "grad_norm": 0.03430556878447533, |
| "kl": 0.03857766184955835, |
| "learning_rate": 1e-06, |
| "loss": 0.0002, |
| "num_tokens": 796632.0, |
| "reward": 0.0, |
| "reward_std": 0.0, |
| "step": 67 |
| }, |
| { |
| "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": 0.08333333333333337, |
| "completions/max_length": 209.0, |
| "completions/max_terminated_length": 209.0, |
| "completions/mean_length": 115.08333333333333, |
| "completions/mean_terminated_length": 125.54545454545455, |
| "completions/min_length": 0.0, |
| "completions/min_terminated_length": 14.0, |
| "epoch": 0.272, |
| "format_failures": 0.0, |
| "grad_norm": 1.6054855585098267, |
| "kl": 0.020691730547696352, |
| "learning_rate": 1e-06, |
| "loss": 0.0511, |
| "num_tokens": 807576.0, |
| "reward": 0.4166666865348816, |
| "reward_std": 0.5149286389350891, |
| "step": 68 |
| }, |
| { |
| "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": 0.08333333333333337, |
| "completions/max_length": 106.0, |
| "completions/max_terminated_length": 106.0, |
| "completions/mean_length": 80.5, |
| "completions/mean_terminated_length": 87.81818181818181, |
| "completions/min_length": 0.0, |
| "completions/min_terminated_length": 64.0, |
| "epoch": 0.276, |
| "format_failures": 0.0, |
| "grad_norm": 1.1459321975708008, |
| "kl": 0.017325148917734623, |
| "learning_rate": 1e-06, |
| "loss": 0.0067, |
| "num_tokens": 814644.0, |
| "reward": 0.0833333358168602, |
| "reward_std": 0.28867512941360474, |
| "step": 69 |
| }, |
| { |
| "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": 0.08333333333333337, |
| "completions/max_length": 304.0, |
| "completions/max_terminated_length": 304.0, |
| "completions/mean_length": 175.58333333333334, |
| "completions/mean_terminated_length": 191.54545454545453, |
| "completions/min_length": 0.0, |
| "completions/min_terminated_length": 17.0, |
| "epoch": 0.28, |
| "format_failures": 0.0, |
| "grad_norm": 0.008818876929581165, |
| "kl": 0.012372600380331278, |
| "learning_rate": 1e-06, |
| "loss": 0.0001, |
| "num_tokens": 826932.0, |
| "reward": 0.0, |
| "reward_std": 0.0, |
| "step": 70 |
| }, |
| { |
| "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": 0.08333333333333337, |
| "completions/max_length": 292.0, |
| "completions/max_terminated_length": 292.0, |
| "completions/mean_length": 195.0, |
| "completions/mean_terminated_length": 212.72727272727272, |
| "completions/min_length": 0.0, |
| "completions/min_terminated_length": 115.0, |
| "epoch": 0.284, |
| "format_failures": 0.0, |
| "grad_norm": 0.014721119776368141, |
| "kl": 0.012880454771220684, |
| "learning_rate": 1e-06, |
| "loss": 0.0001, |
| "num_tokens": 842268.0, |
| "reward": 0.0, |
| "reward_std": 0.0, |
| "step": 71 |
| }, |
| { |
| "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": 0.08333333333333337, |
| "completions/max_length": 145.0, |
| "completions/max_terminated_length": 145.0, |
| "completions/mean_length": 94.83333333333333, |
| "completions/mean_terminated_length": 103.45454545454545, |
| "completions/min_length": 0.0, |
| "completions/min_terminated_length": 75.0, |
| "epoch": 0.288, |
| "format_failures": 0.0, |
| "grad_norm": 0.9220354557037354, |
| "kl": 0.046924193389713764, |
| "learning_rate": 1e-06, |
| "loss": 0.0118, |
| "num_tokens": 849612.0, |
| "reward": 0.0833333358168602, |
| "reward_std": 0.28867512941360474, |
| "step": 72 |
| }, |
| { |
| "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": 0.08333333333333337, |
| "completions/max_length": 262.0, |
| "completions/max_terminated_length": 262.0, |
| "completions/mean_length": 149.5, |
| "completions/mean_terminated_length": 163.0909090909091, |
| "completions/min_length": 0.0, |
| "completions/min_terminated_length": 12.0, |
| "epoch": 0.292, |
| "format_failures": 0.0, |
| "grad_norm": 0.0295345988124609, |
| "kl": 0.03905524965375662, |
| "learning_rate": 1e-06, |
| "loss": 0.0002, |
| "num_tokens": 859632.0, |
| "reward": 0.0, |
| "reward_std": 0.0, |
| "step": 73 |
| }, |
| { |
| "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": 0.08333333333333337, |
| "completions/max_length": 220.0, |
| "completions/max_terminated_length": 220.0, |
| "completions/mean_length": 158.41666666666666, |
| "completions/mean_terminated_length": 172.8181818181818, |
| "completions/min_length": 0.0, |
| "completions/min_terminated_length": 24.0, |
| "epoch": 0.296, |
| "format_failures": 0.0, |
| "grad_norm": 0.11439846456050873, |
| "kl": 0.07962214201688766, |
| "learning_rate": 1e-06, |
| "loss": 0.0004, |
| "num_tokens": 870756.0, |
| "reward": 0.0, |
| "reward_std": 0.0, |
| "step": 74 |
| }, |
| { |
| "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": 0.08333333333333337, |
| "completions/max_length": 223.0, |
| "completions/max_terminated_length": 223.0, |
| "completions/mean_length": 110.08333333333333, |
| "completions/mean_terminated_length": 120.0909090909091, |
| "completions/min_length": 0.0, |
| "completions/min_terminated_length": 11.0, |
| "epoch": 0.3, |
| "format_failures": 0.0, |
| "grad_norm": 0.04706709831953049, |
| "kl": 0.03136777225881815, |
| "learning_rate": 1e-06, |
| "loss": 0.0003, |
| "num_tokens": 887700.0, |
| "reward": 0.0, |
| "reward_std": 0.0, |
| "step": 75 |
| }, |
| { |
| "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": 0.08333333333333337, |
| "completions/max_length": 544.0, |
| "completions/max_terminated_length": 544.0, |
| "completions/mean_length": 219.58333333333334, |
| "completions/mean_terminated_length": 239.54545454545453, |
| "completions/min_length": 0.0, |
| "completions/min_terminated_length": 8.0, |
| "epoch": 0.304, |
| "format_failures": 0.0, |
| "grad_norm": 0.106910839676857, |
| "kl": 0.16153255105018616, |
| "learning_rate": 1e-06, |
| "loss": 0.0004, |
| "num_tokens": 899544.0, |
| "reward": 0.0, |
| "reward_std": 0.0, |
| "step": 76 |
| }, |
| { |
| "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": 0.08333333333333337, |
| "completions/max_length": 175.0, |
| "completions/max_terminated_length": 175.0, |
| "completions/mean_length": 118.66666666666667, |
| "completions/mean_terminated_length": 129.45454545454547, |
| "completions/min_length": 0.0, |
| "completions/min_terminated_length": 48.0, |
| "epoch": 0.308, |
| "format_failures": 0.0, |
| "grad_norm": 0.9582226276397705, |
| "kl": 0.1435188725590706, |
| "learning_rate": 1e-06, |
| "loss": 0.0297, |
| "num_tokens": 909816.0, |
| "reward": 0.1666666716337204, |
| "reward_std": 0.38924944400787354, |
| "step": 77 |
| }, |
| { |
| "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": 0.08333333333333337, |
| "completions/max_length": 200.0, |
| "completions/max_terminated_length": 200.0, |
| "completions/mean_length": 151.83333333333334, |
| "completions/mean_terminated_length": 165.63636363636363, |
| "completions/min_length": 0.0, |
| "completions/min_terminated_length": 85.0, |
| "epoch": 0.312, |
| "format_failures": 0.0, |
| "grad_norm": 0.6430385112762451, |
| "kl": 0.021885435096919537, |
| "learning_rate": 1e-06, |
| "loss": -0.0413, |
| "num_tokens": 919620.0, |
| "reward": 0.0833333358168602, |
| "reward_std": 0.28867512941360474, |
| "step": 78 |
| }, |
| { |
| "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": 0.08333333333333337, |
| "completions/max_length": 127.0, |
| "completions/max_terminated_length": 127.0, |
| "completions/mean_length": 87.66666666666667, |
| "completions/mean_terminated_length": 95.63636363636364, |
| "completions/min_length": 0.0, |
| "completions/min_terminated_length": 72.0, |
| "epoch": 0.316, |
| "format_failures": 0.0, |
| "grad_norm": 0.1316368579864502, |
| "kl": 0.052431097254157066, |
| "learning_rate": 1e-06, |
| "loss": 0.0005, |
| "num_tokens": 930468.0, |
| "reward": 0.0, |
| "reward_std": 0.0, |
| "step": 79 |
| }, |
| { |
| "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": 0.08333333333333337, |
| "completions/max_length": 357.0, |
| "completions/max_terminated_length": 357.0, |
| "completions/mean_length": 188.66666666666666, |
| "completions/mean_terminated_length": 205.8181818181818, |
| "completions/min_length": 0.0, |
| "completions/min_terminated_length": 10.0, |
| "epoch": 0.32, |
| "format_failures": 0.0, |
| "grad_norm": 0.24080750346183777, |
| "kl": 0.25305451452732086, |
| "learning_rate": 1e-06, |
| "loss": 0.0008, |
| "num_tokens": 947112.0, |
| "reward": 0.0, |
| "reward_std": 0.0, |
| "step": 80 |
| }, |
| { |
| "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": 0.08333333333333337, |
| "completions/max_length": 171.0, |
| "completions/max_terminated_length": 171.0, |
| "completions/mean_length": 130.75, |
| "completions/mean_terminated_length": 142.63636363636363, |
| "completions/min_length": 0.0, |
| "completions/min_terminated_length": 77.0, |
| "epoch": 0.324, |
| "format_failures": 0.0, |
| "grad_norm": 0.9585680961608887, |
| "kl": 0.02085646940395236, |
| "learning_rate": 1e-06, |
| "loss": -0.0725, |
| "num_tokens": 956448.0, |
| "reward": 0.5, |
| "reward_std": 0.5222329497337341, |
| "step": 81 |
| }, |
| { |
| "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": 0.08333333333333337, |
| "completions/max_length": 220.0, |
| "completions/max_terminated_length": 220.0, |
| "completions/mean_length": 119.25, |
| "completions/mean_terminated_length": 130.0909090909091, |
| "completions/min_length": 0.0, |
| "completions/min_terminated_length": 11.0, |
| "epoch": 0.328, |
| "format_failures": 0.0, |
| "grad_norm": 0.02760450914502144, |
| "kl": 0.020923216827213764, |
| "learning_rate": 1e-06, |
| "loss": 0.0001, |
| "num_tokens": 966324.0, |
| "reward": 0.0, |
| "reward_std": 0.0, |
| "step": 82 |
| }, |
| { |
| "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": 0.08333333333333337, |
| "completions/max_length": 476.0, |
| "completions/max_terminated_length": 476.0, |
| "completions/mean_length": 252.33333333333334, |
| "completions/mean_terminated_length": 275.27272727272725, |
| "completions/min_length": 0.0, |
| "completions/min_terminated_length": 10.0, |
| "epoch": 0.332, |
| "format_failures": 1.0, |
| "grad_norm": 0.011845018714666367, |
| "kl": 0.017354148440063, |
| "learning_rate": 1e-06, |
| "loss": 0.0001, |
| "num_tokens": 985296.0, |
| "reward": 0.0, |
| "reward_std": 0.0, |
| "step": 83 |
| }, |
| { |
| "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": 0.08333333333333337, |
| "completions/max_length": 234.0, |
| "completions/max_terminated_length": 234.0, |
| "completions/mean_length": 101.16666666666667, |
| "completions/mean_terminated_length": 110.36363636363636, |
| "completions/min_length": 0.0, |
| "completions/min_terminated_length": 10.0, |
| "epoch": 0.336, |
| "format_failures": 1.0, |
| "grad_norm": 0.02075113356113434, |
| "kl": 0.013977942056953907, |
| "learning_rate": 1e-06, |
| "loss": 0.0001, |
| "num_tokens": 998856.0, |
| "reward": 0.0, |
| "reward_std": 0.0, |
| "step": 84 |
| }, |
| { |
| "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": 0.08333333333333337, |
| "completions/max_length": 217.0, |
| "completions/max_terminated_length": 217.0, |
| "completions/mean_length": 156.0, |
| "completions/mean_terminated_length": 170.1818181818182, |
| "completions/min_length": 0.0, |
| "completions/min_terminated_length": 44.0, |
| "epoch": 0.34, |
| "format_failures": 0.0, |
| "grad_norm": 0.018603280186653137, |
| "kl": 0.020112676545977592, |
| "learning_rate": 1e-06, |
| "loss": 0.0001, |
| "num_tokens": 1008864.0, |
| "reward": 0.0, |
| "reward_std": 0.0, |
| "step": 85 |
| }, |
| { |
| "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": 0.08333333333333337, |
| "completions/max_length": 364.0, |
| "completions/max_terminated_length": 364.0, |
| "completions/mean_length": 223.66666666666666, |
| "completions/mean_terminated_length": 244.0, |
| "completions/min_length": 0.0, |
| "completions/min_terminated_length": 20.0, |
| "epoch": 0.344, |
| "format_failures": 0.0, |
| "grad_norm": 0.011895284056663513, |
| "kl": 0.021254747174680233, |
| "learning_rate": 1e-06, |
| "loss": 0.0001, |
| "num_tokens": 1022556.0, |
| "reward": 0.0, |
| "reward_std": 0.0, |
| "step": 86 |
| }, |
| { |
| "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": 0.08333333333333337, |
| "completions/max_length": 159.0, |
| "completions/max_terminated_length": 159.0, |
| "completions/mean_length": 117.66666666666667, |
| "completions/mean_terminated_length": 128.36363636363637, |
| "completions/min_length": 0.0, |
| "completions/min_terminated_length": 112.0, |
| "epoch": 0.348, |
| "format_failures": 0.0, |
| "grad_norm": 1.1451243162155151, |
| "kl": 0.026615198701620102, |
| "learning_rate": 1e-06, |
| "loss": 0.0587, |
| "num_tokens": 1032684.0, |
| "reward": 0.4166666865348816, |
| "reward_std": 0.5149286389350891, |
| "step": 87 |
| }, |
| { |
| "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": 0.08333333333333337, |
| "completions/max_length": 304.0, |
| "completions/max_terminated_length": 304.0, |
| "completions/mean_length": 142.25, |
| "completions/mean_terminated_length": 155.1818181818182, |
| "completions/min_length": 0.0, |
| "completions/min_terminated_length": 12.0, |
| "epoch": 0.352, |
| "format_failures": 2.0, |
| "grad_norm": 0.8502682447433472, |
| "kl": 0.012907921802252531, |
| "learning_rate": 1e-06, |
| "loss": 0.079, |
| "num_tokens": 1067328.0, |
| "reward": 0.1666666716337204, |
| "reward_std": 0.3892494738101959, |
| "step": 88 |
| }, |
| { |
| "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": 0.08333333333333337, |
| "completions/max_length": 177.0, |
| "completions/max_terminated_length": 177.0, |
| "completions/mean_length": 114.5, |
| "completions/mean_terminated_length": 124.9090909090909, |
| "completions/min_length": 0.0, |
| "completions/min_terminated_length": 78.0, |
| "epoch": 0.356, |
| "format_failures": 0.0, |
| "grad_norm": 0.7634170651435852, |
| "kl": 0.08245750516653061, |
| "learning_rate": 1e-06, |
| "loss": 0.0177, |
| "num_tokens": 1074756.0, |
| "reward": 0.0833333358168602, |
| "reward_std": 0.28867512941360474, |
| "step": 89 |
| }, |
| { |
| "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": 0.08333333333333337, |
| "completions/max_length": 141.0, |
| "completions/max_terminated_length": 141.0, |
| "completions/mean_length": 73.0, |
| "completions/mean_terminated_length": 79.63636363636364, |
| "completions/min_length": 0.0, |
| "completions/min_terminated_length": 16.0, |
| "epoch": 0.36, |
| "format_failures": 0.0, |
| "grad_norm": 0.750490665435791, |
| "kl": 0.032081443816423416, |
| "learning_rate": 1e-06, |
| "loss": 0.0503, |
| "num_tokens": 1083096.0, |
| "reward": 0.0833333358168602, |
| "reward_std": 0.28867512941360474, |
| "step": 90 |
| }, |
| { |
| "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": 0.08333333333333337, |
| "completions/max_length": 138.0, |
| "completions/max_terminated_length": 138.0, |
| "completions/mean_length": 53.666666666666664, |
| "completions/mean_terminated_length": 58.54545454545455, |
| "completions/min_length": 0.0, |
| "completions/min_terminated_length": 11.0, |
| "epoch": 0.364, |
| "format_failures": 0.0, |
| "grad_norm": 0.12016791850328445, |
| "kl": 0.04432140104472637, |
| "learning_rate": 1e-06, |
| "loss": 0.0004, |
| "num_tokens": 1090380.0, |
| "reward": 0.0, |
| "reward_std": 0.0, |
| "step": 91 |
| }, |
| { |
| "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": 0.08333333333333337, |
| "completions/max_length": 800.0, |
| "completions/max_terminated_length": 800.0, |
| "completions/mean_length": 397.5, |
| "completions/mean_terminated_length": 433.6363636363636, |
| "completions/min_length": 0.0, |
| "completions/min_terminated_length": 14.0, |
| "epoch": 0.368, |
| "format_failures": 0.0, |
| "grad_norm": 0.012203319929540157, |
| "kl": 0.009247956797480583, |
| "learning_rate": 1e-06, |
| "loss": 0.0, |
| "num_tokens": 1113504.0, |
| "reward": 0.0, |
| "reward_std": 0.0, |
| "step": 92 |
| }, |
| { |
| "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": 0.08333333333333337, |
| "completions/max_length": 261.0, |
| "completions/max_terminated_length": 261.0, |
| "completions/mean_length": 138.75, |
| "completions/mean_terminated_length": 151.36363636363637, |
| "completions/min_length": 0.0, |
| "completions/min_terminated_length": 23.0, |
| "epoch": 0.372, |
| "format_failures": 0.0, |
| "grad_norm": 0.03371990844607353, |
| "kl": 0.029644143767654896, |
| "learning_rate": 1e-06, |
| "loss": 0.0002, |
| "num_tokens": 1125492.0, |
| "reward": 0.0, |
| "reward_std": 0.0, |
| "step": 93 |
| }, |
| { |
| "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": 0.08333333333333337, |
| "completions/max_length": 89.0, |
| "completions/max_terminated_length": 89.0, |
| "completions/mean_length": 51.833333333333336, |
| "completions/mean_terminated_length": 56.54545454545455, |
| "completions/min_length": 0.0, |
| "completions/min_terminated_length": 10.0, |
| "epoch": 0.376, |
| "format_failures": 0.0, |
| "grad_norm": 2.027597427368164, |
| "kl": 0.19823284726589918, |
| "learning_rate": 1e-06, |
| "loss": -0.0712, |
| "num_tokens": 1130748.0, |
| "reward": 0.75, |
| "reward_std": 0.45226702094078064, |
| "step": 94 |
| }, |
| { |
| "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": 0.08333333333333337, |
| "completions/max_length": 118.0, |
| "completions/max_terminated_length": 118.0, |
| "completions/mean_length": 74.33333333333333, |
| "completions/mean_terminated_length": 81.0909090909091, |
| "completions/min_length": 0.0, |
| "completions/min_terminated_length": 62.0, |
| "epoch": 0.38, |
| "format_failures": 0.0, |
| "grad_norm": 1.4458988904953003, |
| "kl": 0.07499337941408157, |
| "learning_rate": 1e-06, |
| "loss": -0.0187, |
| "num_tokens": 1138584.0, |
| "reward": 0.0833333358168602, |
| "reward_std": 0.28867512941360474, |
| "step": 95 |
| }, |
| { |
| "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": 0.08333333333333337, |
| "completions/max_length": 285.0, |
| "completions/max_terminated_length": 285.0, |
| "completions/mean_length": 137.66666666666666, |
| "completions/mean_terminated_length": 150.1818181818182, |
| "completions/min_length": 0.0, |
| "completions/min_terminated_length": 15.0, |
| "epoch": 0.384, |
| "format_failures": 0.0, |
| "grad_norm": 0.03813532739877701, |
| "kl": 0.023914064280688763, |
| "learning_rate": 1e-06, |
| "loss": 0.0002, |
| "num_tokens": 1154124.0, |
| "reward": 0.0, |
| "reward_std": 0.0, |
| "step": 96 |
| }, |
| { |
| "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": 0.08333333333333337, |
| "completions/max_length": 206.0, |
| "completions/max_terminated_length": 206.0, |
| "completions/mean_length": 125.66666666666667, |
| "completions/mean_terminated_length": 137.0909090909091, |
| "completions/min_length": 0.0, |
| "completions/min_terminated_length": 10.0, |
| "epoch": 0.388, |
| "format_failures": 1.0, |
| "grad_norm": 0.016639724373817444, |
| "kl": 0.019042176194489002, |
| "learning_rate": 1e-06, |
| "loss": 0.0001, |
| "num_tokens": 1165560.0, |
| "reward": 0.0, |
| "reward_std": 0.0, |
| "step": 97 |
| }, |
| { |
| "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": 0.08333333333333337, |
| "completions/max_length": 209.0, |
| "completions/max_terminated_length": 209.0, |
| "completions/mean_length": 136.0, |
| "completions/mean_terminated_length": 148.36363636363637, |
| "completions/min_length": 0.0, |
| "completions/min_terminated_length": 11.0, |
| "epoch": 0.392, |
| "format_failures": 0.0, |
| "grad_norm": 0.041289571672677994, |
| "kl": 0.025019565597176552, |
| "learning_rate": 1e-06, |
| "loss": 0.0002, |
| "num_tokens": 1176936.0, |
| "reward": 0.0, |
| "reward_std": 0.0, |
| "step": 98 |
| }, |
| { |
| "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": 0.08333333333333337, |
| "completions/max_length": 539.0, |
| "completions/max_terminated_length": 539.0, |
| "completions/mean_length": 236.75, |
| "completions/mean_terminated_length": 258.27272727272725, |
| "completions/min_length": 0.0, |
| "completions/min_terminated_length": 10.0, |
| "epoch": 0.396, |
| "format_failures": 0.0, |
| "grad_norm": 0.029155507683753967, |
| "kl": 0.03094907756894827, |
| "learning_rate": 1e-06, |
| "loss": 0.0001, |
| "num_tokens": 1194108.0, |
| "reward": 0.0, |
| "reward_std": 0.0, |
| "step": 99 |
| }, |
| { |
| "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": 0.08333333333333337, |
| "completions/max_length": 31.0, |
| "completions/max_terminated_length": 31.0, |
| "completions/mean_length": 21.0, |
| "completions/mean_terminated_length": 22.90909090909091, |
| "completions/min_length": 0.0, |
| "completions/min_terminated_length": 13.0, |
| "epoch": 0.4, |
| "format_failures": 0.0, |
| "grad_norm": 5.876866340637207, |
| "kl": 0.1100139394402504, |
| "learning_rate": 1e-06, |
| "loss": 0.0323, |
| "num_tokens": 1202412.0, |
| "reward": 0.75, |
| "reward_std": 0.45226702094078064, |
| "step": 100 |
| } |
| ], |
| "logging_steps": 1, |
| "max_steps": 1000, |
| "num_input_tokens_seen": 1202412, |
| "num_train_epochs": 4, |
| "save_steps": 500, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": false, |
| "should_training_stop": false |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 0.0, |
| "train_batch_size": 2, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
