End of training

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:480f21ec1837bd316152be3cb8b60e5d205ebe39f275db531e51d1adbed5894e
+oid sha256:0c15a0c4f68c1fe1bdb927d2630bbecdacc9df3895c747019ac5e90948643f9c
 size 73911112

all_results.json

@@ -1,8 +1,8 @@
 {
     "total_flos": 0.0,
-    "train_loss": 7.248813339799653e-07,
-    "train_runtime": 1624.2404,
-    "train_samples": 98,
-    "train_samples_per_second": 0.197,
-    "train_steps_per_second": 0.012
+    "train_loss": 7.633447467014776e-07,
+    "train_runtime": 620.1706,
+    "train_samples": 11,
+    "train_samples_per_second": 0.516,
+    "train_steps_per_second": 0.032
 }

train_results.json

@@ -1,8 +1,8 @@
 {
     "total_flos": 0.0,
-    "train_loss": 7.248813339799653e-07,
-    "train_runtime": 1624.2404,
-    "train_samples": 98,
-    "train_samples_per_second": 0.197,
-    "train_steps_per_second": 0.012
+    "train_loss": 7.633447467014776e-07,
+    "train_runtime": 620.1706,
+    "train_samples": 11,
+    "train_samples_per_second": 0.516,
+    "train_steps_per_second": 0.032
 }

trainer_state.json

@@ -2,7 +2,7 @@
   "best_global_step": null,
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 0.8163265306122449,
+  "epoch": 6.7272727272727275,
   "eval_steps": 500,
   "global_step": 20,
   "is_hyper_param_search": false,
@@ -10,209 +10,209 @@
   "is_world_process_zero": true,
   "log_history": [
     {
-      "completion_length": 347.5625,
-      "epoch": 0.08163265306122448,
-      "grad_norm": 0.49709150195121765,
-      "kl": 0.0,
+      "completion_length": 474.21875,
+      "epoch": 0.7272727272727273,
+      "grad_norm": 11.102519989013672,
+      "kl": 0.0015854837038205005,
       "learning_rate": 5e-07,
       "loss": 0.0,
-      "reward": 1.5538663044571877,
-      "reward_std": 0.8957561552524567,
-      "rewards/concensus_correctness_reward_func": 0.0,
-      "rewards/consensus_reward_func": 0.4375,
+      "reward": 1.165588615462184,
+      "reward_std": 1.2616559863090515,
+      "rewards/concensus_correctness_reward_func": 0.03712499886751175,
+      "rewards/consensus_reward_func": 0.3125,
       "rewards/cumulative_reward_2": 0.0,
-      "rewards/final_correctness_reward_func": 0.1875,
-      "rewards/question_recreation_reward_func": 0.4717725142836571,
+      "rewards/final_correctness_reward_func": 0.25,
+      "rewards/question_recreation_reward_func": 0.4880573768168688,
       "rewards/soft_format_reward_func": 0.0,
-      "rewards/strict_format_reward_func": 0.015625,
-      "rewards/xmlcount_reward_func": 0.44146874733269215,
+      "rewards/strict_format_reward_func": 0.0,
+      "rewards/xmlcount_reward_func": 0.07790625467896461,
       "step": 2
     },
     {
-      "completion_length": 392.34375,
-      "epoch": 0.16326530612244897,
-      "grad_norm": 1.0190365314483643,
-      "kl": 0.00082904764713021,
+      "completion_length": 631.6428571428571,
+      "epoch": 1.3636363636363638,
+      "grad_norm": 0.31869104504585266,
+      "kl": 0.0006388820391813559,
       "learning_rate": 4.864543104251586e-07,
       "loss": 0.0,
-      "reward": 1.1829268708825111,
-      "reward_std": 0.7207849770784378,
-      "rewards/concensus_correctness_reward_func": 0.0078125,
-      "rewards/consensus_reward_func": 0.0625,
+      "reward": 1.0196993138108934,
+      "reward_std": 1.2631095520087652,
+      "rewards/concensus_correctness_reward_func": 0.017857142857142856,
+      "rewards/consensus_reward_func": 0.2857142857142857,
       "rewards/cumulative_reward_2": 0.0,
-      "rewards/final_correctness_reward_func": 0.125,
-      "rewards/question_recreation_reward_func": 0.460333114489913,
+      "rewards/final_correctness_reward_func": 0.0,
+      "rewards/question_recreation_reward_func": 0.564592182636261,
       "rewards/soft_format_reward_func": 0.0,
-      "rewards/strict_format_reward_func": 0.015625,
-      "rewards/xmlcount_reward_func": 0.5116562619805336,
+      "rewards/strict_format_reward_func": 0.0,
+      "rewards/xmlcount_reward_func": 0.15153570952160017,
       "step": 4
     },
     {
-      "completion_length": 404.84375,
-      "epoch": 0.24489795918367346,
-      "grad_norm": 4.023301601409912,
-      "kl": 0.0009292965260101482,
+      "completion_length": 477.92857142857144,
+      "epoch": 2.0,
+      "grad_norm": 0.40221139788627625,
+      "kl": 0.0007509369419754616,
       "learning_rate": 4.472851273490984e-07,
       "loss": 0.0,
-      "reward": 1.3389800153672695,
-      "reward_std": 1.119007632136345,
-      "rewards/concensus_correctness_reward_func": 0.0,
-      "rewards/consensus_reward_func": 0.25,
+      "reward": 1.436357170343399,
+      "reward_std": 0.9000124122415271,
+      "rewards/concensus_correctness_reward_func": 0.11528571162905012,
+      "rewards/consensus_reward_func": 0.21428571428571427,
       "rewards/cumulative_reward_2": 0.0,
-      "rewards/final_correctness_reward_func": 0.125,
-      "rewards/question_recreation_reward_func": 0.6427613086998463,
+      "rewards/final_correctness_reward_func": 0.21428571428571427,
+      "rewards/question_recreation_reward_func": 0.4874285842691149,
       "rewards/soft_format_reward_func": 0.0,
-      "rewards/strict_format_reward_func": 0.046875,
-      "rewards/xmlcount_reward_func": 0.27434375137090683,
+      "rewards/strict_format_reward_func": 0.017857142857142856,
+      "rewards/xmlcount_reward_func": 0.38721428385802675,
       "step": 6
     },
     {
-      "completion_length": 600.53125,
-      "epoch": 0.32653061224489793,
-      "grad_norm": 0.36614489555358887,
-      "kl": 0.0006038047395122703,
+      "completion_length": 482.09375,
+      "epoch": 2.7272727272727275,
+      "grad_norm": 0.4230179786682129,
+      "kl": 0.0007268810477398802,
       "learning_rate": 3.867370395306068e-07,
       "loss": 0.0,
-      "reward": 1.1000889372080564,
-      "reward_std": 1.182983409613371,
-      "rewards/concensus_correctness_reward_func": 0.00787500012665987,
-      "rewards/consensus_reward_func": 0.375,
+      "reward": 0.6160509251058102,
+      "reward_std": 0.9240549877285957,
+      "rewards/concensus_correctness_reward_func": 0.0078125,
+      "rewards/consensus_reward_func": 0.25,
       "rewards/cumulative_reward_2": 0.0,
-      "rewards/final_correctness_reward_func": 0.0625,
-      "rewards/question_recreation_reward_func": 0.7000889368355274,
-      "rewards/soft_format_reward_func": 0.0,
+      "rewards/final_correctness_reward_func": 0.0,
+      "rewards/question_recreation_reward_func": 0.5814571585506201,
+      "rewards/soft_format_reward_func": 0.015625,
       "rewards/strict_format_reward_func": 0.0,
-      "rewards/xmlcount_reward_func": -0.045374996960163116,
+      "rewards/xmlcount_reward_func": -0.23884374648332596,
       "step": 8
     },
     {
-      "completion_length": 436.1875,
-      "epoch": 0.40816326530612246,
-      "grad_norm": 0.6466080546379089,
-      "kl": 0.0008341637658304535,
+      "completion_length": 544.6071428571429,
+      "epoch": 3.3636363636363638,
+      "grad_norm": 0.2655961811542511,
+      "kl": 0.0006771481837079461,
       "learning_rate": 3.1137137178519977e-07,
       "loss": 0.0,
-      "reward": 1.4170492440462112,
-      "reward_std": 0.8611738085746765,
-      "rewards/concensus_correctness_reward_func": 0.0,
-      "rewards/consensus_reward_func": 0.125,
+      "reward": 3.709272067461695,
+      "reward_std": 4.326010116509029,
+      "rewards/concensus_correctness_reward_func": 2.2335714612688338,
+      "rewards/consensus_reward_func": 0.42857142857142855,
       "rewards/cumulative_reward_2": 0.0,
-      "rewards/final_correctness_reward_func": 0.3125,
-      "rewards/question_recreation_reward_func": 0.4472992364317179,
+      "rewards/final_correctness_reward_func": 0.5,
+      "rewards/question_recreation_reward_func": 0.4333436148507254,
       "rewards/soft_format_reward_func": 0.0,
-      "rewards/strict_format_reward_func": 0.015625,
-      "rewards/xmlcount_reward_func": 0.51662502810359,
+      "rewards/strict_format_reward_func": 0.0,
+      "rewards/xmlcount_reward_func": 0.11378570858921323,
       "step": 10
     },
     {
-      "completion_length": 449.71875,
-      "epoch": 0.4897959183673469,
-      "grad_norm": 0.5686748027801514,
-      "kl": 0.0008683982341608498,
+      "completion_length": 455.92857142857144,
+      "epoch": 4.0,
+      "grad_norm": 9.879196166992188,
+      "kl": 0.0007864984716953975,
       "learning_rate": 2.2935516363191693e-07,
       "loss": 0.0,
-      "reward": 1.99321947991848,
-      "reward_std": 1.314138986170292,
-      "rewards/concensus_correctness_reward_func": 0.06006250157952309,
-      "rewards/consensus_reward_func": 0.3125,
+      "reward": 0.733780026435852,
+      "reward_std": 1.1924979175840105,
+      "rewards/concensus_correctness_reward_func": 0.032857143453189304,
+      "rewards/consensus_reward_func": 0.14285714285714285,
       "rewards/cumulative_reward_2": 0.0,
-      "rewards/final_correctness_reward_func": 0.625,
-      "rewards/question_recreation_reward_func": 0.4730007923208177,
+      "rewards/final_correctness_reward_func": 0.0,
+      "rewards/question_recreation_reward_func": 0.4560657377753939,
       "rewards/soft_format_reward_func": 0.0,
-      "rewards/strict_format_reward_func": 0.015625,
-      "rewards/xmlcount_reward_func": 0.507031261920929,
+      "rewards/strict_format_reward_func": 0.0,
+      "rewards/xmlcount_reward_func": 0.10200000235012599,
       "step": 12
     },
     {
-      "completion_length": 390.1875,
-      "epoch": 0.5714285714285714,
-      "grad_norm": 0.4010433554649353,
-      "kl": 0.0008200560114346445,
+      "completion_length": 421.96875,
+      "epoch": 4.7272727272727275,
+      "grad_norm": 0.6101271510124207,
+      "kl": 0.0008735741066629998,
       "learning_rate": 1.4957614383675767e-07,
       "loss": 0.0,
-      "reward": 0.9437324814498425,
-      "reward_std": 0.9435709491372108,
-      "rewards/concensus_correctness_reward_func": 0.03175000101327896,
-      "rewards/consensus_reward_func": 0.0625,
+      "reward": 1.573138004168868,
+      "reward_std": 2.3298346661031246,
+      "rewards/concensus_correctness_reward_func": 0.6328125,
+      "rewards/consensus_reward_func": 0.1875,
       "rewards/cumulative_reward_2": 0.0,
-      "rewards/final_correctness_reward_func": 0.0,
-      "rewards/question_recreation_reward_func": 0.4078262336552143,
-      "rewards/soft_format_reward_func": 0.0,
-      "rewards/strict_format_reward_func": 0.046875,
-      "rewards/xmlcount_reward_func": 0.3947812579572201,
+      "rewards/final_correctness_reward_func": 0.1875,
+      "rewards/question_recreation_reward_func": 0.4569817688316107,
+      "rewards/soft_format_reward_func": 0.015625,
+      "rewards/strict_format_reward_func": 0.015625,
+      "rewards/xmlcount_reward_func": 0.07709374465048313,
       "step": 14
     },
     {
-      "completion_length": 426.9375,
-      "epoch": 0.6530612244897959,
-      "grad_norm": 0.42668047547340393,
-      "kl": 0.0008046694783843122,
+      "completion_length": 398.2142857142857,
+      "epoch": 5.363636363636363,
+      "grad_norm": 0.7817052006721497,
+      "kl": 0.000721364697840597,
       "learning_rate": 8.067960709356478e-08,
       "loss": 0.0,
-      "reward": 1.422092966735363,
-      "reward_std": 1.0166446715593338,
-      "rewards/concensus_correctness_reward_func": 0.0521249994635582,
-      "rewards/consensus_reward_func": 0.25,
+      "reward": 0.9020223702703204,
+      "reward_std": 0.8922735410077232,
+      "rewards/concensus_correctness_reward_func": 0.07528571358748845,
+      "rewards/consensus_reward_func": 0.21428571428571427,
       "rewards/cumulative_reward_2": 0.0,
-      "rewards/final_correctness_reward_func": 0.25,
-      "rewards/question_recreation_reward_func": 0.43712421134114265,
-      "rewards/soft_format_reward_func": 0.015625,
-      "rewards/strict_format_reward_func": 0.046875,
-      "rewards/xmlcount_reward_func": 0.3703437428921461,
+      "rewards/final_correctness_reward_func": 0.0,
+      "rewards/question_recreation_reward_func": 0.3130223538194384,
+      "rewards/soft_format_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.03571428571428571,
+      "rewards/xmlcount_reward_func": 0.26371428796223234,
       "step": 16
     },
     {
-      "completion_length": 496.0625,
-      "epoch": 0.7346938775510204,
-      "grad_norm": 0.315399169921875,
-      "kl": 0.0007264475461852271,
+      "completion_length": 508.82142857142856,
+      "epoch": 6.0,
+      "grad_norm": 0.25807255506515503,
+      "kl": 0.0006166283440377031,
       "learning_rate": 3.013156219837776e-08,
       "loss": 0.0,
-      "reward": 1.3017337992787361,
-      "reward_std": 1.078663855791092,
-      "rewards/concensus_correctness_reward_func": 0.0028750000055879354,
-      "rewards/consensus_reward_func": 0.0625,
+      "reward": 1.0433741807937622,
+      "reward_std": 0.855082631111145,
+      "rewards/concensus_correctness_reward_func": 0.008928571428571428,
+      "rewards/consensus_reward_func": 0.35714285714285715,
       "rewards/cumulative_reward_2": 0.0,
-      "rewards/final_correctness_reward_func": 0.3125,
-      "rewards/question_recreation_reward_func": 0.6079837717115879,
+      "rewards/final_correctness_reward_func": 0.2857142857142857,
+      "rewards/question_recreation_reward_func": 0.4292313264948981,
       "rewards/soft_format_reward_func": 0.0,
-      "rewards/strict_format_reward_func": 0.046875,
-      "rewards/xmlcount_reward_func": 0.269000006839633,
+      "rewards/strict_format_reward_func": 0.0,
+      "rewards/xmlcount_reward_func": -0.03764285360063825,
       "step": 18
     },
     {
-      "completion_length": 399.96875,
-      "epoch": 0.8163265306122449,
-      "grad_norm": 0.5829342007637024,
-      "kl": 0.0007923786251922138,
+      "completion_length": 454.40625,
+      "epoch": 6.7272727272727275,
+      "grad_norm": 0.3790205717086792,
+      "kl": 0.000749875314795645,
       "learning_rate": 3.4096741493194193e-09,
       "loss": 0.0,
-      "reward": 1.70121268928051,
-      "reward_std": 1.1589545384049416,
-      "rewards/concensus_correctness_reward_func": 0.0,
-      "rewards/consensus_reward_func": 0.5,
+      "reward": 0.9435633979737759,
+      "reward_std": 1.0383767876774073,
+      "rewards/concensus_correctness_reward_func": 0.015625,
+      "rewards/consensus_reward_func": 0.1875,
       "rewards/cumulative_reward_2": 0.0,
-      "rewards/final_correctness_reward_func": 0.375,
-      "rewards/question_recreation_reward_func": 0.7043689340353012,
+      "rewards/final_correctness_reward_func": 0.25,
+      "rewards/question_recreation_reward_func": 0.4141883794218302,
       "rewards/soft_format_reward_func": 0.0,
-      "rewards/strict_format_reward_func": 0.0,
-      "rewards/xmlcount_reward_func": 0.12184375338256359,
+      "rewards/strict_format_reward_func": 0.03125,
+      "rewards/xmlcount_reward_func": 0.044999985024333,
       "step": 20
     },
     {
-      "epoch": 0.8163265306122449,
+      "epoch": 6.7272727272727275,
       "step": 20,
       "total_flos": 0.0,
-      "train_loss": 7.248813339799653e-07,
-      "train_runtime": 1624.2404,
-      "train_samples_per_second": 0.197,
-      "train_steps_per_second": 0.012
+      "train_loss": 7.633447467014776e-07,
+      "train_runtime": 620.1706,
+      "train_samples_per_second": 0.516,
+      "train_steps_per_second": 0.032
     }
   ],
   "logging_steps": 2,
   "max_steps": 20,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 1,
+  "num_train_epochs": 10,
   "save_steps": 25,
   "stateful_callbacks": {
     "TrainerControl": {