End of training

README.md

@@ -4,11 +4,11 @@ library_name: transformers
 model_name: Qwen2.5-0.5B-Instruct-Gensyn-Swarm-quick_timid_frog
 tags:
 - generated_from_trainer
-- gensyn
 - trl
-- rl-swarm
-- I am quick timid frog
 - grpo
+- gensyn
+- I am quick timid frog
+- rl-swarm
 licence: license
 ---
 

all_results.json

@@ -1,8 +1,8 @@
 {
     "total_flos": 0.0,
-    "train_loss": 0.0732489401474595,
-    "train_runtime": 1168.9553,
-    "train_samples": 102,
-    "train_samples_per_second": 0.137,
-    "train_steps_per_second": 0.017
+    "train_loss": 0.08094322010874748,
+    "train_runtime": 956.1563,
+    "train_samples": 5,
+    "train_samples_per_second": 0.167,
+    "train_steps_per_second": 0.021
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ffadfd8db278b7139f133097ff2099dd46967944d63636f0cdce3a2ebbfab6a6
+oid sha256:b737bad922ed15c8d1cdc1f0667b18ab7f86f0aa7f0aa8a1198a0cbf7b4e4757
 size 1976163472

train_results.json

@@ -1,8 +1,8 @@
 {
     "total_flos": 0.0,
-    "train_loss": 0.0732489401474595,
-    "train_runtime": 1168.9553,
-    "train_samples": 102,
-    "train_samples_per_second": 0.137,
-    "train_steps_per_second": 0.017
+    "train_loss": 0.08094322010874748,
+    "train_runtime": 956.1563,
+    "train_samples": 5,
+    "train_samples_per_second": 0.167,
+    "train_steps_per_second": 0.021
 }

trainer_state.json

@@ -2,9 +2,9 @@
   "best_global_step": null,
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 0.7843137254901961,
+  "epoch": 9.8,
   "eval_steps": 500,
-  "global_step": 20,
+  "global_step": 10,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -15,38 +15,38 @@
       "clip_ratio/low_mean": 0.0,
       "clip_ratio/low_min": 0.0,
       "clip_ratio/region_mean": 0.0,
-      "completions/clipped_ratio": 0.0625,
-      "completions/max_length": 704.0,
-      "completions/max_terminated_length": 419.5,
-      "completions/mean_length": 199.125,
-      "completions/mean_terminated_length": 145.24107360839844,
-      "completions/min_length": 4.5,
-      "completions/min_terminated_length": 4.5,
-      "epoch": 0.0784313725490196,
+      "completions/clipped_ratio": 0.3125,
+      "completions/max_length": 1024.0,
+      "completions/max_terminated_length": 812.5,
+      "completions/mean_length": 647.75,
+      "completions/mean_terminated_length": 475.3166809082031,
+      "completions/min_length": 138.5,
+      "completions/min_terminated_length": 138.5,
+      "epoch": 1.8,
       "frac_reward_zero_std": 0.0,
-      "grad_norm": 14.074350357055664,
-      "kl": -5.259735091911466e-09,
+      "grad_norm": 3.236729621887207,
+      "kl": 0.00027297700398776215,
       "learning_rate": 5e-07,
-      "loss": -0.0477,
-      "num_tokens": 7282.0,
-      "reward": 0.3504174128174782,
-      "reward_std": 0.37596188485622406,
-      "rewards/concensus_correctness_reward_func/mean": 0.11999999731779099,
-      "rewards/concensus_correctness_reward_func/std": 0.33941125869750977,
-      "rewards/consensus_reward_func/mean": 0.125,
-      "rewards/consensus_reward_func/std": 0.3535533845424652,
+      "loss": -0.001,
+      "num_tokens": 14460.0,
+      "reward": 0.14074324816465378,
+      "reward_std": 0.12783230934292078,
+      "rewards/concensus_correctness_reward_func/mean": 0.07774999737739563,
+      "rewards/concensus_correctness_reward_func/std": 0.21991021931171417,
+      "rewards/consensus_reward_func/mean": 0.0,
+      "rewards/consensus_reward_func/std": 0.0,
       "rewards/cumulative_reward_2/mean": 0.0,
       "rewards/cumulative_reward_2/std": 0.0,
       "rewards/final_correctness_reward_func/mean": 0.0,
       "rewards/final_correctness_reward_func/std": 0.0,
-      "rewards/question_recreation_reward_func/mean": 0.0976049117743969,
-      "rewards/question_recreation_reward_func/std": 0.05009056255221367,
+      "rewards/question_recreation_reward_func/mean": 0.06299325078725815,
+      "rewards/question_recreation_reward_func/std": 0.018077346496284008,
       "rewards/soft_format_reward_func/mean": 0.0,
       "rewards/soft_format_reward_func/std": 0.0,
       "rewards/strict_format_reward_func/mean": 0.0,
       "rewards/strict_format_reward_func/std": 0.0,
-      "rewards/xmlcount_reward_func/mean": 0.0078125,
-      "rewards/xmlcount_reward_func/std": 0.022097086533904076,
+      "rewards/xmlcount_reward_func/mean": 0.0,
+      "rewards/xmlcount_reward_func/std": 0.0,
       "step": 2
     },
     {
@@ -55,32 +55,32 @@
       "clip_ratio/low_mean": 0.0,
       "clip_ratio/low_min": 0.0,
       "clip_ratio/region_mean": 0.0,
-      "completions/clipped_ratio": 0.0,
-      "completions/max_length": 441.0,
-      "completions/max_terminated_length": 441.0,
-      "completions/mean_length": 123.4375,
-      "completions/mean_terminated_length": 123.4375,
-      "completions/min_length": 19.5,
-      "completions/min_terminated_length": 19.5,
-      "epoch": 0.1568627450980392,
+      "completions/clipped_ratio": 0.1875,
+      "completions/max_length": 1024.0,
+      "completions/max_terminated_length": 754.0,
+      "completions/mean_length": 397.25,
+      "completions/mean_terminated_length": 251.08333587646484,
+      "completions/min_length": 4.0,
+      "completions/min_terminated_length": 4.0,
+      "epoch": 3.8,
       "frac_reward_zero_std": 0.0,
-      "grad_norm": 11.524222373962402,
-      "kl": 0.002076731077977456,
+      "grad_norm": 4.27091646194458,
+      "kl": 0.0006203544398886152,
       "learning_rate": 4.864543104251586e-07,
-      "loss": -0.059,
-      "num_tokens": 13353.0,
-      "reward": 0.024305207654833794,
-      "reward_std": 0.012769676744937897,
-      "rewards/concensus_correctness_reward_func/mean": 0.0,
-      "rewards/concensus_correctness_reward_func/std": 0.0,
+      "loss": 0.0127,
+      "num_tokens": 24912.0,
+      "reward": 0.1337380139157176,
+      "reward_std": 0.1742639576550573,
+      "rewards/concensus_correctness_reward_func/mean": 0.11999999731779099,
+      "rewards/concensus_correctness_reward_func/std": 0.33941125869750977,
       "rewards/consensus_reward_func/mean": 0.0,
       "rewards/consensus_reward_func/std": 0.0,
       "rewards/cumulative_reward_2/mean": 0.0,
       "rewards/cumulative_reward_2/std": 0.0,
       "rewards/final_correctness_reward_func/mean": 0.0,
       "rewards/final_correctness_reward_func/std": 0.0,
-      "rewards/question_recreation_reward_func/mean": 0.024305207654833794,
-      "rewards/question_recreation_reward_func/std": 0.012661695014685392,
+      "rewards/question_recreation_reward_func/mean": 0.013738006353378296,
+      "rewards/question_recreation_reward_func/std": 0.010505724931135774,
       "rewards/soft_format_reward_func/mean": 0.0,
       "rewards/soft_format_reward_func/std": 0.0,
       "rewards/strict_format_reward_func/mean": 0.0,
@@ -95,32 +95,32 @@
       "clip_ratio/low_mean": 0.0,
       "clip_ratio/low_min": 0.0,
       "clip_ratio/region_mean": 0.0,
-      "completions/clipped_ratio": 0.3125,
+      "completions/clipped_ratio": 0.1875,
       "completions/max_length": 1024.0,
-      "completions/max_terminated_length": 522.0,
-      "completions/mean_length": 490.875,
-      "completions/mean_terminated_length": 259.14286041259766,
-      "completions/min_length": 54.0,
-      "completions/min_terminated_length": 54.0,
-      "epoch": 0.23529411764705882,
+      "completions/max_terminated_length": 846.5,
+      "completions/mean_length": 528.4375,
+      "completions/mean_terminated_length": 422.8214416503906,
+      "completions/min_length": 4.0,
+      "completions/min_terminated_length": 4.0,
+      "epoch": 5.8,
       "frac_reward_zero_std": 0.0,
-      "grad_norm": 8.319639205932617,
-      "kl": 0.0011255119334236952,
+      "grad_norm": 3.0706632137298584,
+      "kl": 0.0008973750846053008,
       "learning_rate": 4.472851273490984e-07,
-      "loss": 0.1013,
-      "num_tokens": 25303.0,
-      "reward": 0.061401030980050564,
-      "reward_std": 0.022767718299292028,
-      "rewards/concensus_correctness_reward_func/mean": 0.0,
-      "rewards/concensus_correctness_reward_func/std": 0.0,
+      "loss": 0.0403,
+      "num_tokens": 37463.0,
+      "reward": 0.12599835265427828,
+      "reward_std": 0.16678897803649306,
+      "rewards/concensus_correctness_reward_func/mean": 0.1107499971985817,
+      "rewards/concensus_correctness_reward_func/std": 0.31324827671051025,
       "rewards/consensus_reward_func/mean": 0.0,
       "rewards/consensus_reward_func/std": 0.0,
       "rewards/cumulative_reward_2/mean": 0.0,
       "rewards/cumulative_reward_2/std": 0.0,
       "rewards/final_correctness_reward_func/mean": 0.0,
       "rewards/final_correctness_reward_func/std": 0.0,
-      "rewards/question_recreation_reward_func/mean": 0.061401030980050564,
-      "rewards/question_recreation_reward_func/std": 0.025020001688972116,
+      "rewards/question_recreation_reward_func/mean": 0.015248352661728859,
+      "rewards/question_recreation_reward_func/std": 0.015421947930008173,
       "rewards/soft_format_reward_func/mean": 0.0,
       "rewards/soft_format_reward_func/std": 0.0,
       "rewards/strict_format_reward_func/mean": 0.0,
@@ -135,22 +135,22 @@
       "clip_ratio/low_mean": 0.0,
       "clip_ratio/low_min": 0.0,
       "clip_ratio/region_mean": 0.0,
-      "completions/clipped_ratio": 0.125,
+      "completions/clipped_ratio": 0.3125,
       "completions/max_length": 1024.0,
-      "completions/max_terminated_length": 818.0,
-      "completions/mean_length": 428.0,
-      "completions/mean_terminated_length": 342.85716247558594,
-      "completions/min_length": 124.0,
-      "completions/min_terminated_length": 124.0,
-      "epoch": 0.3137254901960784,
+      "completions/max_terminated_length": 744.5,
+      "completions/mean_length": 588.75,
+      "completions/mean_terminated_length": 384.3666687011719,
+      "completions/min_length": 119.0,
+      "completions/min_terminated_length": 119.0,
+      "epoch": 7.8,
       "frac_reward_zero_std": 0.0,
-      "grad_norm": 5.078871250152588,
-      "kl": 0.0014561636453436222,
+      "grad_norm": 3.6464202404022217,
+      "kl": 0.0010676352321752347,
       "learning_rate": 3.867370395306068e-07,
-      "loss": -0.0029,
-      "num_tokens": 36247.0,
-      "reward": 0.028490139171481133,
-      "reward_std": 0.08061825484037399,
+      "loss": 0.1472,
+      "num_tokens": 50979.0,
+      "reward": 0.05322604067623615,
+      "reward_std": 0.02834410360082984,
       "rewards/concensus_correctness_reward_func/mean": 0.0,
       "rewards/concensus_correctness_reward_func/std": 0.0,
       "rewards/consensus_reward_func/mean": 0.0,
@@ -159,135 +159,15 @@
       "rewards/cumulative_reward_2/std": 0.0,
       "rewards/final_correctness_reward_func/mean": 0.0,
       "rewards/final_correctness_reward_func/std": 0.0,
-      "rewards/question_recreation_reward_func/mean": 0.06105264462530613,
-      "rewards/question_recreation_reward_func/std": 0.05342706199735403,
-      "rewards/soft_format_reward_func/mean": 0.0,
-      "rewards/soft_format_reward_func/std": 0.0,
-      "rewards/strict_format_reward_func/mean": 0.0,
-      "rewards/strict_format_reward_func/std": 0.0,
-      "rewards/xmlcount_reward_func/mean": -0.0325625017285347,
-      "rewards/xmlcount_reward_func/std": 0.18048901483416557,
-      "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": 0.0625,
-      "completions/max_length": 815.5,
-      "completions/max_terminated_length": 646.0,
-      "completions/mean_length": 357.5,
-      "completions/mean_terminated_length": 313.2946472167969,
-      "completions/min_length": 61.5,
-      "completions/min_terminated_length": 61.5,
-      "epoch": 0.39215686274509803,
-      "frac_reward_zero_std": 0.0,
-      "grad_norm": 7.4079108238220215,
-      "kl": 0.002010989912378136,
-      "learning_rate": 3.1137137178519977e-07,
-      "loss": 0.032,
-      "num_tokens": 46063.0,
-      "reward": -0.06573907285928726,
-      "reward_std": 0.2942507714033127,
-      "rewards/concensus_correctness_reward_func/mean": 0.0,
-      "rewards/concensus_correctness_reward_func/std": 0.0,
-      "rewards/consensus_reward_func/mean": 0.0,
-      "rewards/consensus_reward_func/std": 0.0,
-      "rewards/cumulative_reward_2/mean": 0.0,
-      "rewards/cumulative_reward_2/std": 0.0,
-      "rewards/final_correctness_reward_func/mean": 0.0,
-      "rewards/final_correctness_reward_func/std": 0.0,
-      "rewards/question_recreation_reward_func/mean": 0.11263591796159744,
-      "rewards/question_recreation_reward_func/std": 0.062143225222826004,
-      "rewards/soft_format_reward_func/mean": 0.0,
-      "rewards/soft_format_reward_func/std": 0.0,
-      "rewards/strict_format_reward_func/mean": 0.0,
-      "rewards/strict_format_reward_func/std": 0.0,
-      "rewards/xmlcount_reward_func/mean": -0.1783750057220459,
-      "rewards/xmlcount_reward_func/std": 0.5045207142829895,
-      "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": 0.125,
-      "completions/max_length": 1024.0,
-      "completions/max_terminated_length": 652.5,
-      "completions/mean_length": 453.0625,
-      "completions/mean_terminated_length": 371.50001525878906,
-      "completions/min_length": 142.5,
-      "completions/min_terminated_length": 142.5,
-      "epoch": 0.47058823529411764,
-      "frac_reward_zero_std": 0.0,
-      "grad_norm": 4.9817328453063965,
-      "kl": 0.001734087803924922,
-      "learning_rate": 2.2935516363191693e-07,
-      "loss": 0.1614,
-      "num_tokens": 57408.0,
-      "reward": 0.02063838578760624,
-      "reward_std": 0.005743567133322358,
-      "rewards/concensus_correctness_reward_func/mean": 0.0,
-      "rewards/concensus_correctness_reward_func/std": 0.0,
-      "rewards/consensus_reward_func/mean": 0.0,
-      "rewards/consensus_reward_func/std": 0.0,
-      "rewards/cumulative_reward_2/mean": 0.0,
-      "rewards/cumulative_reward_2/std": 0.0,
-      "rewards/final_correctness_reward_func/mean": 0.0,
-      "rewards/final_correctness_reward_func/std": 0.0,
-      "rewards/question_recreation_reward_func/mean": 0.02063838578760624,
-      "rewards/question_recreation_reward_func/std": 0.010399988386780024,
+      "rewards/question_recreation_reward_func/mean": 0.05322604067623615,
+      "rewards/question_recreation_reward_func/std": 0.03214400727301836,
       "rewards/soft_format_reward_func/mean": 0.0,
       "rewards/soft_format_reward_func/std": 0.0,
       "rewards/strict_format_reward_func/mean": 0.0,
       "rewards/strict_format_reward_func/std": 0.0,
       "rewards/xmlcount_reward_func/mean": 0.0,
       "rewards/xmlcount_reward_func/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": 0.0625,
-      "completions/max_length": 721.0,
-      "completions/max_terminated_length": 657.5,
-      "completions/mean_length": 304.625,
-      "completions/mean_terminated_length": 265.2589416503906,
-      "completions/min_length": 8.5,
-      "completions/min_terminated_length": 8.5,
-      "epoch": 0.5490196078431373,
-      "frac_reward_zero_std": 0.0,
-      "grad_norm": 5.698163986206055,
-      "kl": 0.004235875936501543,
-      "learning_rate": 1.4957614383675767e-07,
-      "loss": 0.2616,
-      "num_tokens": 66378.0,
-      "reward": 0.11159846559166908,
-      "reward_std": 0.03269663080573082,
-      "rewards/concensus_correctness_reward_func/mean": 0.0,
-      "rewards/concensus_correctness_reward_func/std": 0.0,
-      "rewards/consensus_reward_func/mean": 0.0,
-      "rewards/consensus_reward_func/std": 0.0,
-      "rewards/cumulative_reward_2/mean": 0.0,
-      "rewards/cumulative_reward_2/std": 0.0,
-      "rewards/final_correctness_reward_func/mean": 0.0,
-      "rewards/final_correctness_reward_func/std": 0.0,
-      "rewards/question_recreation_reward_func/mean": 0.11159846559166908,
-      "rewards/question_recreation_reward_func/std": 0.043566348031163216,
-      "rewards/soft_format_reward_func/mean": 0.0,
-      "rewards/soft_format_reward_func/std": 0.0,
-      "rewards/strict_format_reward_func/mean": 0.0,
-      "rewards/strict_format_reward_func/std": 0.0,
-      "rewards/xmlcount_reward_func/mean": 0.0,
-      "rewards/xmlcount_reward_func/std": 0.0,
-      "step": 14
+      "step": 8
     },
     {
       "clip_ratio/high_max": 0.0,
@@ -295,62 +175,22 @@
       "clip_ratio/low_mean": 0.0,
       "clip_ratio/low_min": 0.0,
       "clip_ratio/region_mean": 0.0,
-      "completions/clipped_ratio": 0.1875,
+      "completions/clipped_ratio": 0.375,
       "completions/max_length": 1024.0,
-      "completions/max_terminated_length": 648.0,
-      "completions/mean_length": 489.125,
-      "completions/mean_terminated_length": 362.08335876464844,
-      "completions/min_length": 145.5,
-      "completions/min_terminated_length": 145.5,
-      "epoch": 0.6274509803921569,
-      "frac_reward_zero_std": 0.0,
-      "grad_norm": 4.048318386077881,
-      "kl": 0.001558408792334376,
-      "learning_rate": 8.067960709356478e-08,
-      "loss": 0.0805,
-      "num_tokens": 78300.0,
-      "reward": 0.020350982900708914,
-      "reward_std": 0.004954211530275643,
-      "rewards/concensus_correctness_reward_func/mean": 0.0,
-      "rewards/concensus_correctness_reward_func/std": 0.0,
-      "rewards/consensus_reward_func/mean": 0.0,
-      "rewards/consensus_reward_func/std": 0.0,
-      "rewards/cumulative_reward_2/mean": 0.0,
-      "rewards/cumulative_reward_2/std": 0.0,
-      "rewards/final_correctness_reward_func/mean": 0.0,
-      "rewards/final_correctness_reward_func/std": 0.0,
-      "rewards/question_recreation_reward_func/mean": 0.020350982900708914,
-      "rewards/question_recreation_reward_func/std": 0.018837051233276725,
-      "rewards/soft_format_reward_func/mean": 0.0,
-      "rewards/soft_format_reward_func/std": 0.0,
-      "rewards/strict_format_reward_func/mean": 0.0,
-      "rewards/strict_format_reward_func/std": 0.0,
-      "rewards/xmlcount_reward_func/mean": 0.0,
-      "rewards/xmlcount_reward_func/std": 0.0,
-      "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": 0.0,
-      "completions/max_length": 569.0,
-      "completions/max_terminated_length": 569.0,
-      "completions/mean_length": 310.4375,
-      "completions/mean_terminated_length": 310.4375,
-      "completions/min_length": 57.5,
-      "completions/min_terminated_length": 57.5,
-      "epoch": 0.7058823529411765,
+      "completions/max_terminated_length": 817.5,
+      "completions/mean_length": 669.4375,
+      "completions/mean_terminated_length": 470.0,
+      "completions/min_length": 148.5,
+      "completions/min_terminated_length": 148.5,
+      "epoch": 9.8,
       "frac_reward_zero_std": 0.0,
-      "grad_norm": 5.309350490570068,
-      "kl": 0.0033653288937784964,
-      "learning_rate": 3.013156219837776e-08,
-      "loss": 0.0653,
-      "num_tokens": 87363.0,
-      "reward": 0.032109410502016544,
-      "reward_std": 0.014090773183852434,
+      "grad_norm": 3.48903226852417,
+      "kl": 0.0014760679550818168,
+      "learning_rate": 3.1137137178519977e-07,
+      "loss": 0.2054,
+      "num_tokens": 65786.0,
+      "reward": 0.011943170800805092,
+      "reward_std": 0.0064218416810035706,
       "rewards/concensus_correctness_reward_func/mean": 0.0,
       "rewards/concensus_correctness_reward_func/std": 0.0,
       "rewards/consensus_reward_func/mean": 0.0,
@@ -359,70 +199,30 @@
       "rewards/cumulative_reward_2/std": 0.0,
       "rewards/final_correctness_reward_func/mean": 0.0,
       "rewards/final_correctness_reward_func/std": 0.0,
-      "rewards/question_recreation_reward_func/mean": 0.032109410502016544,
-      "rewards/question_recreation_reward_func/std": 0.026289107277989388,
+      "rewards/question_recreation_reward_func/mean": 0.011943170800805092,
+      "rewards/question_recreation_reward_func/std": 0.0063162968726828694,
       "rewards/soft_format_reward_func/mean": 0.0,
       "rewards/soft_format_reward_func/std": 0.0,
       "rewards/strict_format_reward_func/mean": 0.0,
       "rewards/strict_format_reward_func/std": 0.0,
       "rewards/xmlcount_reward_func/mean": 0.0,
       "rewards/xmlcount_reward_func/std": 0.0,
-      "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": 0.0,
-      "completions/max_length": 462.5,
-      "completions/max_terminated_length": 462.5,
-      "completions/mean_length": 204.375,
-      "completions/mean_terminated_length": 204.375,
-      "completions/min_length": 67.0,
-      "completions/min_terminated_length": 67.0,
-      "epoch": 0.7843137254901961,
-      "frac_reward_zero_std": 0.0,
-      "grad_norm": 7.5336384773254395,
-      "kl": 0.003989799035480246,
-      "learning_rate": 3.4096741493194193e-09,
-      "loss": 0.1399,
-      "num_tokens": 94729.0,
-      "reward": 0.27750399708747864,
-      "reward_std": 0.015088059706613421,
-      "rewards/concensus_correctness_reward_func/mean": 0.0,
-      "rewards/concensus_correctness_reward_func/std": 0.0,
-      "rewards/consensus_reward_func/mean": 0.0,
-      "rewards/consensus_reward_func/std": 0.0,
-      "rewards/cumulative_reward_2/mean": 0.0,
-      "rewards/cumulative_reward_2/std": 0.0,
-      "rewards/final_correctness_reward_func/mean": 0.25,
-      "rewards/final_correctness_reward_func/std": 0.4629100561141968,
-      "rewards/question_recreation_reward_func/mean": 0.02587897516787052,
-      "rewards/question_recreation_reward_func/std": 0.0247750636190176,
-      "rewards/soft_format_reward_func/mean": 0.0,
-      "rewards/soft_format_reward_func/std": 0.0,
-      "rewards/strict_format_reward_func/mean": 0.0,
-      "rewards/strict_format_reward_func/std": 0.0,
-      "rewards/xmlcount_reward_func/mean": 0.0016250000335276127,
-      "rewards/xmlcount_reward_func/std": 0.004596194718033075,
-      "step": 20
+      "step": 10
     },
     {
-      "epoch": 0.7843137254901961,
-      "step": 20,
+      "epoch": 9.8,
+      "step": 10,
       "total_flos": 0.0,
-      "train_loss": 0.0732489401474595,
-      "train_runtime": 1168.9553,
-      "train_samples_per_second": 0.137,
-      "train_steps_per_second": 0.017
+      "train_loss": 0.08094322010874748,
+      "train_runtime": 956.1563,
+      "train_samples_per_second": 0.167,
+      "train_steps_per_second": 0.021
     }
   ],
   "logging_steps": 2,
   "max_steps": 20,
-  "num_input_tokens_seen": 94729,
-  "num_train_epochs": 1,
+  "num_input_tokens_seen": 65786,
+  "num_train_epochs": 10,
   "save_steps": 25,
   "stateful_callbacks": {
     "TrainerControl": {
@@ -430,8 +230,8 @@
         "should_epoch_stop": false,
         "should_evaluate": false,
         "should_log": false,
-        "should_save": true,
-        "should_training_stop": true
+        "should_save": false,
+        "should_training_stop": false
       },
       "attributes": {}
     }