End of training

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0634e2f97a968ff8ee0c30a4e51b0c6780a895d817342ac26a4e2d6e2dd6e576
+oid sha256:0ef875dfc91d14df97712e673b6783804ecc245f2a88ecb2fd8d8f4f12b3751e
 size 842289128

all_results.json

@@ -1,8 +1,8 @@
 {
     "total_flos": 0.0,
-    "train_loss": 6.820050475653261e-06,
-    "train_runtime": 4436.3893,
+    "train_loss": 1.4550274863722734e-05,
+    "train_runtime": 4069.2289,
     "train_samples": 9,
-    "train_samples_per_second": 0.018,
+    "train_samples_per_second": 0.02,
     "train_steps_per_second": 0.002
 }

train_results.json

@@ -1,8 +1,8 @@
 {
     "total_flos": 0.0,
-    "train_loss": 6.820050475653261e-06,
-    "train_runtime": 4436.3893,
+    "train_loss": 1.4550274863722734e-05,
+    "train_runtime": 4069.2289,
     "train_samples": 9,
-    "train_samples_per_second": 0.018,
+    "train_samples_per_second": 0.02,
     "train_steps_per_second": 0.002
 }

trainer_state.json

@@ -10,107 +10,107 @@
   "is_world_process_zero": true,
   "log_history": [
     {
-      "completion_length": 342.375,
+      "completion_length": 306.6875,
       "epoch": 0.8888888888888888,
-      "grad_norm": 0.4052090644836426,
-      "kl": 0.012759620556607842,
+      "grad_norm": 0.39643576741218567,
+      "kl": 0.017112158122472465,
       "learning_rate": 5e-07,
       "loss": 0.0,
-      "reward": 5.321024656295776,
-      "reward_std": 0.10023243329487741,
-      "rewards/concensus_correctness_reward_func": 1.3215000201016665,
+      "reward": 5.813093453645706,
+      "reward_std": 0.14963281876407564,
+      "rewards/concensus_correctness_reward_func": 1.7742499868036248,
       "rewards/consensus_reward_func": 2.0,
       "rewards/cumulative_reward_2": 0.0,
       "rewards/final_correctness_reward_func": 1.0,
-      "rewards/question_recreation_reward_func": 0.9986496642231941,
+      "rewards/question_recreation_reward_func": 0.9020308256149292,
       "rewards/soft_format_reward_func": 0.0,
       "rewards/strict_format_reward_func": 0.0,
-      "rewards/xmlcount_reward_func": 0.0008750008419156075,
+      "rewards/xmlcount_reward_func": 0.1368125006556511,
       "step": 2
     },
     {
-      "completion_length": 301.3,
+      "completion_length": 305.1,
       "epoch": 1.4444444444444444,
-      "grad_norm": 0.41138553619384766,
-      "kl": 0.01866710316389799,
+      "grad_norm": 0.4071302115917206,
+      "kl": 0.024027385748922825,
       "learning_rate": 4.415111107797445e-07,
       "loss": 0.0,
-      "reward": 6.361130905151367,
-      "reward_std": 0.054022852703928946,
-      "rewards/concensus_correctness_reward_func": 2.0604000210762026,
+      "reward": 6.344730997085572,
+      "reward_std": 0.05741699002683163,
+      "rewards/concensus_correctness_reward_func": 2.0456000403501093,
       "rewards/consensus_reward_func": 2.0,
       "rewards/cumulative_reward_2": 0.0,
       "rewards/final_correctness_reward_func": 1.2,
       "rewards/question_recreation_reward_func": 0.9993311047554017,
       "rewards/soft_format_reward_func": 0.0,
       "rewards/strict_format_reward_func": 0.0,
-      "rewards/xmlcount_reward_func": 0.10139999985694885,
+      "rewards/xmlcount_reward_func": 0.09979999661445618,
       "step": 4
     },
     {
-      "completion_length": 351.8,
+      "completion_length": 301.2,
       "epoch": 2.0,
-      "grad_norm": 0.16249226033687592,
-      "kl": 0.01092723784968257,
+      "grad_norm": 0.14513656497001648,
+      "kl": 0.018292231671512126,
       "learning_rate": 2.934120444167326e-07,
       "loss": 0.0,
-      "reward": 5.599708366394043,
-      "reward_std": 0.058265522867441175,
-      "rewards/concensus_correctness_reward_func": 1.3724000126123428,
+      "reward": 6.5120086669921875,
+      "reward_std": 0.07509479415602982,
+      "rewards/concensus_correctness_reward_func": 2.1467999637126924,
       "rewards/consensus_reward_func": 2.0,
       "rewards/cumulative_reward_2": 0.0,
       "rewards/final_correctness_reward_func": 1.2,
       "rewards/question_recreation_reward_func": 0.998508358001709,
       "rewards/soft_format_reward_func": 0.0,
       "rewards/strict_format_reward_func": 0.0,
-      "rewards/xmlcount_reward_func": 0.028800001740455626,
+      "rewards/xmlcount_reward_func": 0.16669999957084655,
       "step": 6
     },
     {
-      "completion_length": 346.8125,
+      "completion_length": 315.5625,
       "epoch": 2.888888888888889,
-      "grad_norm": 0.31399527192115784,
-      "kl": 0.014027125318534672,
+      "grad_norm": 0.36357581615448,
+      "kl": 0.018164650886319578,
       "learning_rate": 1.2500000000000005e-07,
       "loss": 0.0,
-      "reward": 5.361516207456589,
-      "reward_std": 0.12613017641706392,
-      "rewards/concensus_correctness_reward_func": 1.3687500227242708,
+      "reward": 5.899516224861145,
+      "reward_std": 0.0996137821348384,
+      "rewards/concensus_correctness_reward_func": 1.8205000023008324,
       "rewards/consensus_reward_func": 2.0,
       "rewards/cumulative_reward_2": 0.0,
       "rewards/final_correctness_reward_func": 1.0,
       "rewards/question_recreation_reward_func": 0.9989537969231606,
       "rewards/soft_format_reward_func": 0.0,
       "rewards/strict_format_reward_func": 0.0,
-      "rewards/xmlcount_reward_func": -0.0061874995008111,
+      "rewards/xmlcount_reward_func": 0.08006250020116568,
       "step": 8
     },
     {
-      "completion_length": 316.2,
+      "completion_length": 302.2,
       "epoch": 3.4444444444444446,
-      "grad_norm": 0.42416098713874817,
-      "kl": 0.014391851425170899,
+      "grad_norm": 0.4360595941543579,
+      "kl": 0.019211999699473382,
       "learning_rate": 1.507684480352292e-08,
-      "loss": -0.0,
-      "reward": 6.528926610946655,
-      "reward_std": 0.06547806216403843,
-      "rewards/concensus_correctness_reward_func": 1.8683999925851822,
+      "loss": 0.0,
+      "reward": 7.314918279647827,
+      "reward_std": 0.11413912773132324,
+      "rewards/concensus_correctness_reward_func": 2.593599981069565,
       "rewards/consensus_reward_func": 2.0,
       "rewards/cumulative_reward_2": 0.0,
       "rewards/final_correctness_reward_func": 1.6,
-      "rewards/question_recreation_reward_func": 0.9990267753601074,
+      "rewards/question_recreation_reward_func": 0.9948183536529541,
       "rewards/soft_format_reward_func": 0.0,
       "rewards/strict_format_reward_func": 0.0,
-      "rewards/xmlcount_reward_func": 0.06150000095367432,
+      "rewards/xmlcount_reward_func": 0.12649999856948851,
       "step": 10
     },
     {
       "epoch": 3.4444444444444446,
       "step": 10,
       "total_flos": 0.0,
-      "train_loss": 6.820050475653261e-06,
-      "train_runtime": 4436.3893,
-      "train_samples_per_second": 0.018,
+      "train_loss": 1.4550274863722734e-05,
+      "train_runtime": 4069.2289,
+      "train_samples_per_second": 0.02,
       "train_steps_per_second": 0.002
     }
   ],