End of training

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:24de2d863463f7b28fb15734c33945d7e982587ae810850fc4b2ac6f78d980b9
+oid sha256:29d26ece11bdc68ac15009b5d8af102934b8be7c845f35fe26dc23d9b40c811c
 size 35237104

all_results.json

@@ -1,8 +1,8 @@
 {
     "total_flos": 0.0,
-    "train_loss": 3.4262249300809343e-06,
-    "train_runtime": 1179.869,
-    "train_samples": 20,
-    "train_samples_per_second": 0.271,
-    "train_steps_per_second": 0.017
+    "train_loss": 2.5033270617313973e-06,
+    "train_runtime": 561.0387,
+    "train_samples": 14,
+    "train_samples_per_second": 0.57,
+    "train_steps_per_second": 0.036
 }

train_results.json

@@ -1,8 +1,8 @@
 {
     "total_flos": 0.0,
-    "train_loss": 3.4262249300809343e-06,
-    "train_runtime": 1179.869,
-    "train_samples": 20,
-    "train_samples_per_second": 0.271,
-    "train_steps_per_second": 0.017
+    "train_loss": 2.5033270617313973e-06,
+    "train_runtime": 561.0387,
+    "train_samples": 14,
+    "train_samples_per_second": 0.57,
+    "train_steps_per_second": 0.036
 }

trainer_state.json

@@ -1,7 +1,7 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 4.0,
+  "epoch": 5.0,
   "eval_steps": 500,
   "global_step": 20,
   "is_hyper_param_search": false,
@@ -9,209 +9,209 @@
   "is_world_process_zero": true,
   "log_history": [
     {
-      "completion_length": 385.78125,
-      "epoch": 0.4,
-      "grad_norm": 1.8336645364761353,
-      "kl": 0.001763462569215335,
+      "completion_length": 380.75,
+      "epoch": 0.5714285714285714,
+      "grad_norm": 3.2658472061157227,
+      "kl": 0.002441615804855246,
       "learning_rate": 4.965903258506806e-07,
       "loss": 0.0,
-      "reward": 0.7109862742945552,
-      "reward_std": 1.3441396206617355,
-      "rewards/concensus_correctness_reward_func": 0.06012500077486038,
+      "reward": 0.25457921205088496,
+      "reward_std": 0.7497397037222981,
+      "rewards/concensus_correctness_reward_func": 0.0,
       "rewards/consensus_reward_func": 0.0,
       "rewards/cumulative_reward_2": 0.0,
-      "rewards/final_correctness_reward_func": 0.25,
-      "rewards/question_recreation_reward_func": 0.3131737904623151,
+      "rewards/final_correctness_reward_func": 0.1875,
+      "rewards/question_recreation_reward_func": 0.17123546241782606,
       "rewards/soft_format_reward_func": 0.0,
       "rewards/strict_format_reward_func": 0.0,
-      "rewards/xmlcount_reward_func": 0.08768750261515379,
+      "rewards/xmlcount_reward_func": -0.10415624268352985,
       "step": 2
     },
     {
-      "completion_length": 379.03125,
-      "epoch": 0.8,
-      "grad_norm": 3.063690423965454,
-      "kl": 0.0015668635605834424,
+      "completion_length": 477.5416666666667,
+      "epoch": 1.0,
+      "grad_norm": 9.751885414123535,
+      "kl": 0.0027688652432213226,
       "learning_rate": 4.698684378016222e-07,
       "loss": 0.0,
-      "reward": 1.4691463224589825,
-      "reward_std": 1.5921650007367134,
-      "rewards/concensus_correctness_reward_func": 0.1197500005364418,
-      "rewards/consensus_reward_func": 0.1875,
+      "reward": 0.2258941854039828,
+      "reward_std": 0.5309189086159071,
+      "rewards/concensus_correctness_reward_func": 0.0,
+      "rewards/consensus_reward_func": 0.0,
       "rewards/cumulative_reward_2": 0.0,
-      "rewards/final_correctness_reward_func": 0.4375,
-      "rewards/question_recreation_reward_func": 0.3682400956749916,
+      "rewards/final_correctness_reward_func": 0.08333333333333333,
+      "rewards/question_recreation_reward_func": 0.22968583926558495,
       "rewards/soft_format_reward_func": 0.0,
       "rewards/strict_format_reward_func": 0.0,
-      "rewards/xmlcount_reward_func": 0.35615625558421016,
+      "rewards/xmlcount_reward_func": -0.08712499557683866,
       "step": 4
     },
     {
-      "completion_length": 405.375,
-      "epoch": 1.2,
-      "grad_norm": 1.3559259176254272,
-      "kl": 0.00803125137463212,
+      "completion_length": 340.28125,
+      "epoch": 1.5714285714285714,
+      "grad_norm": 2.0237159729003906,
+      "kl": 0.0016068490367615595,
       "learning_rate": 4.193203929064353e-07,
       "loss": 0.0,
-      "reward": 1.1987021006643772,
-      "reward_std": 1.1934575475752354,
-      "rewards/concensus_correctness_reward_func": 0.019687499850988388,
-      "rewards/consensus_reward_func": 0.25,
+      "reward": 0.41925548017024994,
+      "reward_std": 0.694330379832536,
+      "rewards/concensus_correctness_reward_func": 0.0078125,
+      "rewards/consensus_reward_func": 0.0,
       "rewards/cumulative_reward_2": 0.0,
-      "rewards/final_correctness_reward_func": 0.375,
-      "rewards/question_recreation_reward_func": 0.3701396007090807,
-      "rewards/soft_format_reward_func": 0.015625,
+      "rewards/final_correctness_reward_func": 0.1875,
+      "rewards/question_recreation_reward_func": 0.13991173182148486,
+      "rewards/soft_format_reward_func": 0.0,
       "rewards/strict_format_reward_func": 0.0,
-      "rewards/xmlcount_reward_func": 0.16825000056996942,
+      "rewards/xmlcount_reward_func": 0.08403124753385782,
       "step": 6
     },
     {
-      "completion_length": 326.40625,
-      "epoch": 1.6,
-      "grad_norm": 7.137951374053955,
-      "kl": 0.002187835576478392,
+      "completion_length": 340.1666666666667,
+      "epoch": 2.0,
+      "grad_norm": 2.802516460418701,
+      "kl": 0.00871574851529052,
       "learning_rate": 3.5042385616324236e-07,
       "loss": 0.0,
-      "reward": 1.200685321353376,
-      "reward_std": 1.6040349267423153,
-      "rewards/concensus_correctness_reward_func": 0.15181250125169754,
-      "rewards/consensus_reward_func": 0.125,
+      "reward": 2.3500575547417006,
+      "reward_std": 2.4711769918600717,
+      "rewards/concensus_correctness_reward_func": 1.6666666666666667,
+      "rewards/consensus_reward_func": 0.0,
       "rewards/cumulative_reward_2": 0.0,
-      "rewards/final_correctness_reward_func": 0.375,
-      "rewards/question_recreation_reward_func": 0.4635916016995907,
-      "rewards/soft_format_reward_func": 0.015625,
+      "rewards/final_correctness_reward_func": 0.3333333333333333,
+      "rewards/question_recreation_reward_func": 0.36718254288037616,
+      "rewards/soft_format_reward_func": 0.0,
       "rewards/strict_format_reward_func": 0.0,
-      "rewards/xmlcount_reward_func": 0.06965625332668424,
+      "rewards/xmlcount_reward_func": -0.017124999935428303,
       "step": 8
     },
     {
-      "completion_length": 263.3125,
-      "epoch": 2.0,
-      "grad_norm": 11.289843559265137,
-      "kl": 0.002439464231429156,
+      "completion_length": 412.78125,
+      "epoch": 2.571428571428571,
+      "grad_norm": 1.4599255323410034,
+      "kl": 0.0019848650772473775,
       "learning_rate": 2.706448363680831e-07,
       "loss": 0.0,
-      "reward": 1.0160155966877937,
-      "reward_std": 1.113827582448721,
+      "reward": 0.2086572889238596,
+      "reward_std": 0.4712572921998799,
       "rewards/concensus_correctness_reward_func": 0.0,
       "rewards/consensus_reward_func": 0.0,
       "rewards/cumulative_reward_2": 0.0,
-      "rewards/final_correctness_reward_func": 0.375,
-      "rewards/question_recreation_reward_func": 0.3645468205213547,
+      "rewards/final_correctness_reward_func": 0.0625,
+      "rewards/question_recreation_reward_func": 0.2022197861224413,
       "rewards/soft_format_reward_func": 0.0,
-      "rewards/strict_format_reward_func": 0.015625,
-      "rewards/xmlcount_reward_func": 0.26084375102072954,
+      "rewards/strict_format_reward_func": 0.0,
+      "rewards/xmlcount_reward_func": -0.05606249812990427,
       "step": 10
     },
     {
-      "completion_length": 318.96875,
-      "epoch": 2.4,
-      "grad_norm": 3.153017044067383,
-      "kl": 0.0032806367307784967,
+      "completion_length": 506.5416666666667,
+      "epoch": 3.0,
+      "grad_norm": 2.3391497135162354,
+      "kl": 0.0016113731738490362,
       "learning_rate": 1.886286282148002e-07,
       "loss": 0.0,
-      "reward": 1.0621484704315662,
-      "reward_std": 0.9562191925942898,
-      "rewards/concensus_correctness_reward_func": 0.023499999195337296,
-      "rewards/consensus_reward_func": 0.125,
+      "reward": 1.0101854726672173,
+      "reward_std": 2.4752619365851083,
+      "rewards/concensus_correctness_reward_func": 0.8333333333333334,
+      "rewards/consensus_reward_func": 0.0,
       "rewards/cumulative_reward_2": 0.0,
-      "rewards/final_correctness_reward_func": 0.3125,
-      "rewards/question_recreation_reward_func": 0.4058672459796071,
+      "rewards/final_correctness_reward_func": 0.16666666666666666,
+      "rewards/question_recreation_reward_func": 0.25510215759277344,
       "rewards/soft_format_reward_func": 0.0,
       "rewards/strict_format_reward_func": 0.0,
-      "rewards/xmlcount_reward_func": 0.19528124667704105,
+      "rewards/xmlcount_reward_func": -0.244916670024395,
       "step": 12
     },
     {
-      "completion_length": 387.09375,
-      "epoch": 2.8,
-      "grad_norm": 4.154686450958252,
-      "kl": 0.0036658041208283976,
+      "completion_length": 472.40625,
+      "epoch": 3.571428571428571,
+      "grad_norm": 1.8769646883010864,
+      "kl": 0.001703827321762219,
       "learning_rate": 1.1326296046939333e-07,
       "loss": 0.0,
-      "reward": 0.7330729253590107,
-      "reward_std": 1.2864641938358545,
-      "rewards/concensus_correctness_reward_func": 0.15856249630451202,
-      "rewards/consensus_reward_func": 0.125,
+      "reward": 1.6477174372412264,
+      "reward_std": 1.7785091185942292,
+      "rewards/concensus_correctness_reward_func": 1.2578125,
+      "rewards/consensus_reward_func": 0.0625,
       "rewards/cumulative_reward_2": 0.0,
-      "rewards/final_correctness_reward_func": 0.25,
-      "rewards/question_recreation_reward_func": 0.3413854278624058,
+      "rewards/final_correctness_reward_func": 0.1875,
+      "rewards/question_recreation_reward_func": 0.34727994725108147,
       "rewards/soft_format_reward_func": 0.0,
       "rewards/strict_format_reward_func": 0.0,
-      "rewards/xmlcount_reward_func": -0.14187500346451998,
+      "rewards/xmlcount_reward_func": -0.20737499929964542,
       "step": 14
     },
     {
-      "completion_length": 369.5,
-      "epoch": 3.2,
-      "grad_norm": 3.3538591861724854,
-      "kl": 0.005900904878217261,
+      "completion_length": 371.0833333333333,
+      "epoch": 4.0,
+      "grad_norm": 4.798888683319092,
+      "kl": 0.004414637068596979,
       "learning_rate": 5.271487265090163e-08,
       "loss": 0.0,
-      "reward": 1.0797511721029878,
-      "reward_std": 0.8328760787844658,
-      "rewards/concensus_correctness_reward_func": 0.0,
-      "rewards/consensus_reward_func": 0.0625,
+      "reward": 1.1725354778269927,
+      "reward_std": 2.4872060331205526,
+      "rewards/concensus_correctness_reward_func": 0.8333333333333334,
+      "rewards/consensus_reward_func": 0.0,
       "rewards/cumulative_reward_2": 0.0,
-      "rewards/final_correctness_reward_func": 0.4375,
-      "rewards/question_recreation_reward_func": 0.4021574263460934,
+      "rewards/final_correctness_reward_func": 0.16666666666666666,
+      "rewards/question_recreation_reward_func": 0.18761878398557505,
       "rewards/soft_format_reward_func": 0.0,
       "rewards/strict_format_reward_func": 0.0,
-      "rewards/xmlcount_reward_func": 0.1775937546044588,
+      "rewards/xmlcount_reward_func": -0.015083340307076773,
       "step": 16
     },
     {
-      "completion_length": 305.59375,
-      "epoch": 3.6,
-      "grad_norm": 1.6157883405685425,
-      "kl": 0.003512895797030069,
+      "completion_length": 457.0,
+      "epoch": 4.571428571428571,
+      "grad_norm": 2.0245349407196045,
+      "kl": 0.0014776990574318916,
       "learning_rate": 1.3545689574841341e-08,
       "loss": 0.0,
-      "reward": 1.1209254786372185,
-      "reward_std": 1.3270590491592884,
-      "rewards/concensus_correctness_reward_func": 0.05974999815225601,
-      "rewards/consensus_reward_func": 0.0625,
+      "reward": 0.26492203027009964,
+      "reward_std": 0.8164641298353672,
+      "rewards/concensus_correctness_reward_func": 0.0,
+      "rewards/consensus_reward_func": 0.0,
       "rewards/cumulative_reward_2": 0.0,
-      "rewards/final_correctness_reward_func": 0.5625,
-      "rewards/question_recreation_reward_func": 0.3955192044377327,
-      "rewards/soft_format_reward_func": 0.015625,
+      "rewards/final_correctness_reward_func": 0.1875,
+      "rewards/question_recreation_reward_func": 0.23551576770842075,
+      "rewards/soft_format_reward_func": 0.0,
       "rewards/strict_format_reward_func": 0.0,
-      "rewards/xmlcount_reward_func": 0.025031255558133125,
+      "rewards/xmlcount_reward_func": -0.15809375420212746,
       "step": 18
     },
     {
-      "completion_length": 359.84375,
-      "epoch": 4.0,
-      "grad_norm": 6.384725093841553,
-      "kl": 0.0019049076654482633,
+      "completion_length": 284.0,
+      "epoch": 5.0,
+      "grad_norm": 77.61033630371094,
+      "kl": 0.0036686273330512145,
       "learning_rate": 0.0,
       "loss": 0.0,
-      "reward": 0.6302484348416328,
-      "reward_std": 1.1225376315414906,
-      "rewards/concensus_correctness_reward_func": 0.019375000149011612,
-      "rewards/consensus_reward_func": 0.125,
+      "reward": 0.7023521810770035,
+      "reward_std": 0.8045435522993406,
+      "rewards/concensus_correctness_reward_func": 0.03675000121196111,
+      "rewards/consensus_reward_func": 0.08333333333333333,
       "rewards/cumulative_reward_2": 0.0,
       "rewards/final_correctness_reward_func": 0.25,
-      "rewards/question_recreation_reward_func": 0.30777968652546406,
+      "rewards/question_recreation_reward_func": 0.11218550615012646,
       "rewards/soft_format_reward_func": 0.0,
       "rewards/strict_format_reward_func": 0.0,
-      "rewards/xmlcount_reward_func": -0.07190625183284283,
+      "rewards/xmlcount_reward_func": 0.2200833410024643,
       "step": 20
     },
     {
-      "epoch": 4.0,
+      "epoch": 5.0,
       "step": 20,
       "total_flos": 0.0,
-      "train_loss": 3.4262249300809343e-06,
-      "train_runtime": 1179.869,
-      "train_samples_per_second": 0.271,
-      "train_steps_per_second": 0.017
+      "train_loss": 2.5033270617313973e-06,
+      "train_runtime": 561.0387,
+      "train_samples_per_second": 0.57,
+      "train_steps_per_second": 0.036
     }
   ],
   "logging_steps": 2,
   "max_steps": 20,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 4,
+  "num_train_epochs": 7,
   "save_steps": 25,
   "stateful_callbacks": {
     "TrainerControl": {