End of training

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:33f4a9a080cdb24709a045192feae7bb251edd9119ed80ae3a544c5d292f5809
+oid sha256:c2f6ce216bf587af85e89f212099466d2699f5f0256ff7a22350016c50459b30
 size 35237104

all_results.json

@@ -1,8 +1,8 @@
 {
     "total_flos": 0.0,
-    "train_loss": 2.704163705402607e-06,
-    "train_runtime": 2165.7434,
+    "train_loss": 5.493214257512591e-06,
+    "train_runtime": 1348.1291,
     "train_samples": 11,
-    "train_samples_per_second": 0.148,
-    "train_steps_per_second": 0.009
+    "train_samples_per_second": 0.237,
+    "train_steps_per_second": 0.015
 }

train_results.json

@@ -1,8 +1,8 @@
 {
     "total_flos": 0.0,
-    "train_loss": 2.704163705402607e-06,
-    "train_runtime": 2165.7434,
+    "train_loss": 5.493214257512591e-06,
+    "train_runtime": 1348.1291,
     "train_samples": 11,
-    "train_samples_per_second": 0.148,
-    "train_steps_per_second": 0.009
+    "train_samples_per_second": 0.237,
+    "train_steps_per_second": 0.015
 }

trainer_state.json

@@ -10,203 +10,203 @@
   "is_world_process_zero": true,
   "log_history": [
     {
-      "completion_length": 299.5625,
+      "completion_length": 309.8125,
       "epoch": 0.7272727272727273,
-      "grad_norm": 1.944827914237976,
-      "kl": 0.0013179814268369228,
+      "grad_norm": 8.192431449890137,
+      "kl": 0.0013196841391618364,
       "learning_rate": 4.965903258506806e-07,
       "loss": 0.0,
-      "reward": 0.11043636966496706,
-      "reward_std": 0.20629275910323486,
+      "reward": 0.22969480406027287,
+      "reward_std": 0.39862296264618635,
       "rewards/concensus_correctness_reward_func": 0.0,
       "rewards/consensus_reward_func": 0.0,
       "rewards/cumulative_reward_2": 0.0,
-      "rewards/final_correctness_reward_func": 0.0,
-      "rewards/question_recreation_reward_func": 0.08568636968266219,
+      "rewards/final_correctness_reward_func": 0.125,
+      "rewards/question_recreation_reward_func": 0.09497605240903795,
       "rewards/soft_format_reward_func": 0.0,
       "rewards/strict_format_reward_func": 0.0,
-      "rewards/xmlcount_reward_func": 0.024750000797212124,
+      "rewards/xmlcount_reward_func": 0.00971875200048089,
       "step": 2
     },
     {
-      "completion_length": 397.0357142857143,
+      "completion_length": 334.5,
       "epoch": 1.3636363636363638,
-      "grad_norm": 3.8321359157562256,
-      "kl": 0.0012097215512767434,
+      "grad_norm": 16.709056854248047,
+      "kl": 0.01644311550106587,
       "learning_rate": 4.698684378016222e-07,
       "loss": 0.0,
-      "reward": 0.09989675240857261,
-      "reward_std": 0.3151250896709306,
-      "rewards/concensus_correctness_reward_func": 0.0,
+      "reward": 1.084287445460047,
+      "reward_std": 1.6699770393648319,
+      "rewards/concensus_correctness_reward_func": 0.7142857142857143,
       "rewards/consensus_reward_func": 0.0,
       "rewards/cumulative_reward_2": 0.0,
-      "rewards/final_correctness_reward_func": 0.0,
-      "rewards/question_recreation_reward_func": 0.10703961631017071,
+      "rewards/final_correctness_reward_func": 0.14285714285714285,
+      "rewards/question_recreation_reward_func": 0.17485884497208254,
       "rewards/soft_format_reward_func": 0.0,
       "rewards/strict_format_reward_func": 0.0,
-      "rewards/xmlcount_reward_func": -0.00714285671710968,
+      "rewards/xmlcount_reward_func": 0.05228571646979877,
       "step": 4
     },
     {
-      "completion_length": 269.7142857142857,
+      "completion_length": 345.17857142857144,
       "epoch": 2.0,
-      "grad_norm": 1.893737554550171,
-      "kl": 0.0021692034040045527,
+      "grad_norm": 2.4254114627838135,
+      "kl": 0.001126350327727518,
       "learning_rate": 4.193203929064353e-07,
       "loss": 0.0,
-      "reward": 0.10630716649549347,
-      "reward_std": 0.2536022617880787,
+      "reward": 0.16946781054139137,
+      "reward_std": 0.30478086562028955,
       "rewards/concensus_correctness_reward_func": 0.0,
       "rewards/consensus_reward_func": 0.0,
       "rewards/cumulative_reward_2": 0.0,
       "rewards/final_correctness_reward_func": 0.0,
-      "rewards/question_recreation_reward_func": 0.13895002433231898,
+      "rewards/question_recreation_reward_func": 0.1678249617772443,
       "rewards/soft_format_reward_func": 0.0,
       "rewards/strict_format_reward_func": 0.0,
-      "rewards/xmlcount_reward_func": -0.03264285836900983,
+      "rewards/xmlcount_reward_func": 0.0016428555494972638,
       "step": 6
     },
     {
-      "completion_length": 412.15625,
+      "completion_length": 426.03125,
       "epoch": 2.7272727272727275,
-      "grad_norm": 24.022533416748047,
-      "kl": 0.004952997565851547,
+      "grad_norm": 1.3110556602478027,
+      "kl": 0.0015403831857838668,
       "learning_rate": 3.5042385616324236e-07,
       "loss": 0.0,
-      "reward": 0.14648076053708792,
-      "reward_std": 0.18240744748618454,
+      "reward": 0.26408953219652176,
+      "reward_std": 0.27848068880848587,
       "rewards/concensus_correctness_reward_func": 0.0,
       "rewards/consensus_reward_func": 0.0,
       "rewards/cumulative_reward_2": 0.0,
       "rewards/final_correctness_reward_func": 0.0,
-      "rewards/question_recreation_reward_func": 0.07976200641132891,
+      "rewards/question_recreation_reward_func": 0.18487078556790948,
       "rewards/soft_format_reward_func": 0.0,
       "rewards/strict_format_reward_func": 0.0,
-      "rewards/xmlcount_reward_func": 0.06671875299070962,
+      "rewards/xmlcount_reward_func": 0.07921875081956387,
       "step": 8
     },
     {
-      "completion_length": 320.82142857142856,
+      "completion_length": 334.10714285714283,
       "epoch": 3.3636363636363638,
-      "grad_norm": 18.309104919433594,
-      "kl": 0.001723618620807039,
+      "grad_norm": 1.4304771423339844,
+      "kl": 0.0011949114850722253,
       "learning_rate": 2.706448363680831e-07,
       "loss": 0.0,
-      "reward": 0.28387557928051266,
-      "reward_std": 0.3226744021688189,
+      "reward": 0.2872551094208445,
+      "reward_std": 0.42002125616584507,
       "rewards/concensus_correctness_reward_func": 0.0,
       "rewards/consensus_reward_func": 0.0,
       "rewards/cumulative_reward_2": 0.0,
       "rewards/final_correctness_reward_func": 0.0,
-      "rewards/question_recreation_reward_func": 0.1718755800809179,
+      "rewards/question_recreation_reward_func": 0.18396939922656333,
       "rewards/soft_format_reward_func": 0.0,
       "rewards/strict_format_reward_func": 0.0,
-      "rewards/xmlcount_reward_func": 0.11199999707085746,
+      "rewards/xmlcount_reward_func": 0.1032857107264655,
       "step": 10
     },
     {
-      "completion_length": 336.4642857142857,
+      "completion_length": 290.07142857142856,
       "epoch": 4.0,
-      "grad_norm": 4.879215240478516,
-      "kl": 0.001175968292435365,
+      "grad_norm": 35.479034423828125,
+      "kl": 0.020025783434643278,
       "learning_rate": 1.886286282148002e-07,
       "loss": 0.0,
-      "reward": 0.23535038105079106,
-      "reward_std": 0.2977558132261038,
+      "reward": 0.2351770829409361,
+      "reward_std": 0.317613242992333,
       "rewards/concensus_correctness_reward_func": 0.0,
-      "rewards/consensus_reward_func": 0.07142857142857142,
+      "rewards/consensus_reward_func": 0.0,
       "rewards/cumulative_reward_2": 0.0,
       "rewards/final_correctness_reward_func": 0.0,
-      "rewards/question_recreation_reward_func": 0.09688608508024897,
+      "rewards/question_recreation_reward_func": 0.12906994510974204,
       "rewards/soft_format_reward_func": 0.0,
       "rewards/strict_format_reward_func": 0.0,
-      "rewards/xmlcount_reward_func": 0.06703571336609977,
+      "rewards/xmlcount_reward_func": 0.10610714235476085,
       "step": 12
     },
     {
-      "completion_length": 314.8125,
+      "completion_length": 385.1875,
       "epoch": 4.7272727272727275,
-      "grad_norm": 21.388477325439453,
-      "kl": 0.009420267095265444,
+      "grad_norm": 1.7368310689926147,
+      "kl": 0.0016509672495885752,
       "learning_rate": 1.1326296046939333e-07,
       "loss": 0.0,
-      "reward": 0.21050509018823504,
-      "reward_std": 0.20859105559065938,
+      "reward": 0.14453346747905016,
+      "reward_std": 0.6091575883328915,
       "rewards/concensus_correctness_reward_func": 0.0,
       "rewards/consensus_reward_func": 0.0,
       "rewards/cumulative_reward_2": 0.0,
-      "rewards/final_correctness_reward_func": 0.0,
-      "rewards/question_recreation_reward_func": 0.14122384111396968,
+      "rewards/final_correctness_reward_func": 0.25,
+      "rewards/question_recreation_reward_func": 0.179158469196409,
       "rewards/soft_format_reward_func": 0.0,
       "rewards/strict_format_reward_func": 0.0,
-      "rewards/xmlcount_reward_func": 0.06928124953992665,
+      "rewards/xmlcount_reward_func": -0.2846250347793102,
       "step": 14
     },
     {
-      "completion_length": 384.5357142857143,
+      "completion_length": 329.92857142857144,
       "epoch": 5.363636363636363,
-      "grad_norm": 5.775881767272949,
-      "kl": 0.0031449103594890664,
+      "grad_norm": 16.084075927734375,
+      "kl": 0.00911175856266969,
       "learning_rate": 5.271487265090163e-08,
       "loss": 0.0,
-      "reward": 0.07761396547513348,
-      "reward_std": 0.16731864054288184,
+      "reward": 0.08588471157210213,
+      "reward_std": 0.1177412923425436,
       "rewards/concensus_correctness_reward_func": 0.0,
       "rewards/consensus_reward_func": 0.0,
       "rewards/cumulative_reward_2": 0.0,
       "rewards/final_correctness_reward_func": 0.0,
-      "rewards/question_recreation_reward_func": 0.09357824894998755,
+      "rewards/question_recreation_reward_func": 0.047313284394996505,
       "rewards/soft_format_reward_func": 0.0,
       "rewards/strict_format_reward_func": 0.0,
-      "rewards/xmlcount_reward_func": -0.015964286667960032,
+      "rewards/xmlcount_reward_func": 0.038571428773658614,
       "step": 16
     },
     {
-      "completion_length": 294.35714285714283,
+      "completion_length": 295.35714285714283,
       "epoch": 6.0,
-      "grad_norm": 1.587579369544983,
-      "kl": 0.0016204850814704383,
+      "grad_norm": 27.753393173217773,
+      "kl": 0.006975149270147085,
       "learning_rate": 1.3545689574841341e-08,
       "loss": 0.0,
-      "reward": 0.21252788177558354,
-      "reward_std": 0.163387374154159,
+      "reward": 0.4774811161415918,
+      "reward_std": 0.5152423062494823,
       "rewards/concensus_correctness_reward_func": 0.0,
       "rewards/consensus_reward_func": 0.0,
       "rewards/cumulative_reward_2": 0.0,
-      "rewards/final_correctness_reward_func": 0.0,
-      "rewards/question_recreation_reward_func": 0.1455278785101005,
+      "rewards/final_correctness_reward_func": 0.14285714285714285,
+      "rewards/question_recreation_reward_func": 0.15815968705075129,
       "rewards/soft_format_reward_func": 0.0,
       "rewards/strict_format_reward_func": 0.0,
-      "rewards/xmlcount_reward_func": 0.06699999954019274,
+      "rewards/xmlcount_reward_func": 0.17646428516932897,
       "step": 18
     },
     {
-      "completion_length": 451.5625,
+      "completion_length": 244.34375,
       "epoch": 6.7272727272727275,
-      "grad_norm": 14.947336196899414,
-      "kl": 0.001650080863328185,
+      "grad_norm": 13.965374946594238,
+      "kl": 0.00237850533449091,
       "learning_rate": 0.0,
       "loss": 0.0,
-      "reward": 0.1783973490819335,
-      "reward_std": 0.26468708156608045,
-      "rewards/concensus_correctness_reward_func": 0.0,
-      "rewards/consensus_reward_func": 0.0,
+      "reward": 0.6199594191275537,
+      "reward_std": 0.5588421156862751,
+      "rewards/concensus_correctness_reward_func": 0.09943749941885471,
+      "rewards/consensus_reward_func": 0.0625,
       "rewards/cumulative_reward_2": 0.0,
-      "rewards/final_correctness_reward_func": 0.0,
-      "rewards/question_recreation_reward_func": 0.13683485565707088,
+      "rewards/final_correctness_reward_func": 0.125,
+      "rewards/question_recreation_reward_func": 0.19514692039228976,
       "rewards/soft_format_reward_func": 0.0,
       "rewards/strict_format_reward_func": 0.0,
-      "rewards/xmlcount_reward_func": 0.04156249761581421,
+      "rewards/xmlcount_reward_func": 0.137874998152256,
       "step": 20
     },
     {
       "epoch": 6.7272727272727275,
       "step": 20,
       "total_flos": 0.0,
-      "train_loss": 2.704163705402607e-06,
-      "train_runtime": 2165.7434,
-      "train_samples_per_second": 0.148,
-      "train_steps_per_second": 0.009
+      "train_loss": 5.493214257512591e-06,
+      "train_runtime": 1348.1291,
+      "train_samples_per_second": 0.237,
+      "train_steps_per_second": 0.015
     }
   ],
   "logging_steps": 2,