End of training

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:90a4634d4e32d8bae324fc9fcee63aff9f8f3335033675070f6751ffca84b7ce
+oid sha256:f9b2cea86f052ae6665697fc4d8835b4f0c9e02b29529451bd6259d6d4eccc95
 size 73911112

all_results.json

@@ -1,8 +1,8 @@
 {
     "total_flos": 0.0,
-    "train_loss": 7.482587704998878e-07,
-    "train_runtime": 902.2542,
-    "train_samples": 18,
-    "train_samples_per_second": 0.355,
-    "train_steps_per_second": 0.022
+    "train_loss": 6.444601922339643e-07,
+    "train_runtime": 1608.3431,
+    "train_samples": 11,
+    "train_samples_per_second": 0.199,
+    "train_steps_per_second": 0.012
 }

train_results.json

@@ -1,8 +1,8 @@
 {
     "total_flos": 0.0,
-    "train_loss": 7.482587704998878e-07,
-    "train_runtime": 902.2542,
-    "train_samples": 18,
-    "train_samples_per_second": 0.355,
-    "train_steps_per_second": 0.022
+    "train_loss": 6.444601922339643e-07,
+    "train_runtime": 1608.3431,
+    "train_samples": 11,
+    "train_samples_per_second": 0.199,
+    "train_steps_per_second": 0.012
 }

trainer_state.json

@@ -2,7 +2,7 @@
   "best_global_step": null,
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 4.0,
+  "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": 453.40625,
-      "epoch": 0.4444444444444444,
-      "grad_norm": 0.48038437962532043,
-      "kl": 0.0006780548646929674,
+      "completion_length": 467.90625,
+      "epoch": 0.7272727272727273,
+      "grad_norm": 0.485609233379364,
+      "kl": 0.0007106798075255938,
       "learning_rate": 5e-07,
       "loss": 0.0,
-      "reward": 1.69500782340765,
-      "reward_std": 1.3618378229439259,
-      "rewards/concensus_correctness_reward_func": 0.2198750041425228,
-      "rewards/consensus_reward_func": 0.5,
+      "reward": 1.2300263158977032,
+      "reward_std": 1.1693743616342545,
+      "rewards/concensus_correctness_reward_func": 0.0,
+      "rewards/consensus_reward_func": 0.4375,
       "rewards/cumulative_reward_2": 0.0,
-      "rewards/final_correctness_reward_func": 0.1875,
-      "rewards/question_recreation_reward_func": 0.49538280069828033,
+      "rewards/final_correctness_reward_func": 0.0,
+      "rewards/question_recreation_reward_func": 0.6245888117700815,
       "rewards/soft_format_reward_func": 0.0,
       "rewards/strict_format_reward_func": 0.0,
-      "rewards/xmlcount_reward_func": 0.29225000366568565,
+      "rewards/xmlcount_reward_func": 0.16793749667704105,
       "step": 2
     },
     {
-      "completion_length": 440.0625,
-      "epoch": 0.8888888888888888,
-      "grad_norm": 0.48385727405548096,
-      "kl": 0.0007756603299640119,
+      "completion_length": 461.39285714285717,
+      "epoch": 1.3636363636363638,
+      "grad_norm": 0.4931257367134094,
+      "kl": 0.0007476569485983678,
       "learning_rate": 4.864543104251586e-07,
       "loss": 0.0,
-      "reward": 1.4402526542544365,
-      "reward_std": 1.5317449048161507,
-      "rewards/concensus_correctness_reward_func": 0.023375000804662704,
-      "rewards/consensus_reward_func": 0.6875,
+      "reward": 0.2825267655508859,
+      "reward_std": 1.0751322656869888,
+      "rewards/concensus_correctness_reward_func": 0.0,
+      "rewards/consensus_reward_func": 0.14285714285714285,
       "rewards/cumulative_reward_2": 0.0,
       "rewards/final_correctness_reward_func": 0.0,
-      "rewards/question_recreation_reward_func": 0.5139088854193687,
-      "rewards/soft_format_reward_func": 0.015625,
-      "rewards/strict_format_reward_func": 0.015625,
-      "rewards/xmlcount_reward_func": 0.1842187475413084,
+      "rewards/question_recreation_reward_func": 0.3599196289266859,
+      "rewards/soft_format_reward_func": 0.0,
+      "rewards/strict_format_reward_func": 0.0,
+      "rewards/xmlcount_reward_func": -0.22024999346051896,
       "step": 4
     },
     {
-      "completion_length": 263.375,
-      "epoch": 1.2222222222222223,
-      "grad_norm": 3.26983380317688,
-      "kl": 0.0018498417290781315,
+      "completion_length": 568.2142857142857,
+      "epoch": 2.0,
+      "grad_norm": 0.22118231654167175,
+      "kl": 0.000631775295395138,
       "learning_rate": 4.472851273490984e-07,
       "loss": 0.0,
-      "reward": 2.162368575731913,
-      "reward_std": 1.4121030271053314,
-      "rewards/concensus_correctness_reward_func": 0.10649999976158142,
-      "rewards/consensus_reward_func": 0.75,
+      "reward": 1.604168461901801,
+      "reward_std": 1.2598716957228524,
+      "rewards/concensus_correctness_reward_func": 0.054285715733255656,
+      "rewards/consensus_reward_func": 0.6428571428571429,
       "rewards/cumulative_reward_2": 0.0,
       "rewards/final_correctness_reward_func": 0.0,
-      "rewards/question_recreation_reward_func": 0.6925769150257111,
+      "rewards/question_recreation_reward_func": 0.5840612916009766,
       "rewards/soft_format_reward_func": 0.0,
-      "rewards/strict_format_reward_func": 0.020833333333333332,
-      "rewards/xmlcount_reward_func": 0.5924583474795023,
+      "rewards/strict_format_reward_func": 0.017857142857142856,
+      "rewards/xmlcount_reward_func": 0.30510713905096054,
       "step": 6
     },
     {
-      "completion_length": 311.375,
-      "epoch": 1.6666666666666665,
-      "grad_norm": 0.713490903377533,
-      "kl": 0.0007732290396234021,
+      "completion_length": 471.34375,
+      "epoch": 2.7272727272727275,
+      "grad_norm": 0.4210323393344879,
+      "kl": 0.0007207773451227695,
       "learning_rate": 3.867370395306068e-07,
       "loss": 0.0,
-      "reward": 2.291649080812931,
-      "reward_std": 1.4728753510862589,
-      "rewards/concensus_correctness_reward_func": 0.20025000721216202,
-      "rewards/consensus_reward_func": 0.9375,
+      "reward": 1.0432898998260498,
+      "reward_std": 0.9795822128653526,
+      "rewards/concensus_correctness_reward_func": 0.0,
+      "rewards/consensus_reward_func": 0.3125,
       "rewards/cumulative_reward_2": 0.0,
-      "rewards/final_correctness_reward_func": 0.1875,
-      "rewards/question_recreation_reward_func": 0.5669302828609943,
-      "rewards/soft_format_reward_func": 0.0,
+      "rewards/final_correctness_reward_func": 0.0,
+      "rewards/question_recreation_reward_func": 0.4576649102382362,
+      "rewards/soft_format_reward_func": 0.015625,
       "rewards/strict_format_reward_func": 0.0,
-      "rewards/xmlcount_reward_func": 0.39946875255554914,
+      "rewards/xmlcount_reward_func": 0.2575000021606684,
       "step": 8
     },
     {
-      "completion_length": 614.9166666666666,
-      "epoch": 2.0,
-      "grad_norm": 0.23068159818649292,
-      "kl": 0.0005923892022110522,
+      "completion_length": 477.5357142857143,
+      "epoch": 3.3636363636363638,
+      "grad_norm": 0.36557313799858093,
+      "kl": 0.0007496414140665106,
       "learning_rate": 3.1137137178519977e-07,
       "loss": 0.0,
-      "reward": 1.028176764647166,
-      "reward_std": 0.8733372464776039,
-      "rewards/concensus_correctness_reward_func": 0.0,
-      "rewards/consensus_reward_func": 0.25,
+      "reward": 0.5420836593423571,
+      "reward_std": 0.947016784123012,
+      "rewards/concensus_correctness_reward_func": 0.02257142961025238,
+      "rewards/consensus_reward_func": 0.14285714285714285,
       "rewards/cumulative_reward_2": 0.0,
       "rewards/final_correctness_reward_func": 0.0,
-      "rewards/question_recreation_reward_func": 0.4852600619196892,
-      "rewards/soft_format_reward_func": 0.020833333333333332,
+      "rewards/question_recreation_reward_func": 0.37286937875407083,
+      "rewards/soft_format_reward_func": 0.0,
       "rewards/strict_format_reward_func": 0.0,
-      "rewards/xmlcount_reward_func": 0.2720833346247673,
+      "rewards/xmlcount_reward_func": 0.003785711313996996,
       "step": 10
     },
     {
-      "completion_length": 542.03125,
-      "epoch": 2.4444444444444446,
-      "grad_norm": 0.29893001914024353,
-      "kl": 0.0006597758474526927,
+      "completion_length": 544.0714285714286,
+      "epoch": 4.0,
+      "grad_norm": 0.4929836690425873,
+      "kl": 0.0007356132014787623,
       "learning_rate": 2.2935516363191693e-07,
       "loss": 0.0,
-      "reward": 1.8402126878499985,
-      "reward_std": 0.9209578447043896,
-      "rewards/concensus_correctness_reward_func": 0.3070000112056732,
-      "rewards/consensus_reward_func": 0.75,
+      "reward": 1.1885977642876762,
+      "reward_std": 1.199118435382843,
+      "rewards/concensus_correctness_reward_func": 0.0,
+      "rewards/consensus_reward_func": 0.42857142857142855,
       "rewards/cumulative_reward_2": 0.0,
-      "rewards/final_correctness_reward_func": 0.25,
-      "rewards/question_recreation_reward_func": 0.5566814253106713,
+      "rewards/final_correctness_reward_func": 0.0,
+      "rewards/question_recreation_reward_func": 0.496383473277092,
       "rewards/soft_format_reward_func": 0.0,
       "rewards/strict_format_reward_func": 0.0,
-      "rewards/xmlcount_reward_func": -0.023468753322958946,
+      "rewards/xmlcount_reward_func": 0.26364285392420633,
       "step": 12
     },
     {
-      "completion_length": 370.09375,
-      "epoch": 2.888888888888889,
-      "grad_norm": 0.6849364042282104,
-      "kl": 0.0008487552186124958,
+      "completion_length": 503.9375,
+      "epoch": 4.7272727272727275,
+      "grad_norm": 0.4013039171695709,
+      "kl": 0.0006397504039341584,
       "learning_rate": 1.4957614383675767e-07,
       "loss": 0.0,
-      "reward": 1.28958485648036,
-      "reward_std": 1.0341310314834118,
-      "rewards/concensus_correctness_reward_func": 0.0,
-      "rewards/consensus_reward_func": 0.375,
+      "reward": 0.9641091949306428,
+      "reward_std": 0.9753384664654732,
+      "rewards/concensus_correctness_reward_func": 0.05925000458955765,
+      "rewards/consensus_reward_func": 0.4375,
       "rewards/cumulative_reward_2": 0.0,
       "rewards/final_correctness_reward_func": 0.0,
-      "rewards/question_recreation_reward_func": 0.5932410396635532,
+      "rewards/question_recreation_reward_func": 0.397171713411808,
       "rewards/soft_format_reward_func": 0.0,
-      "rewards/strict_format_reward_func": 0.015625,
-      "rewards/xmlcount_reward_func": 0.3057187721133232,
+      "rewards/strict_format_reward_func": 0.0,
+      "rewards/xmlcount_reward_func": 0.07018749974668026,
       "step": 14
     },
     {
-      "completion_length": 426.5416666666667,
-      "epoch": 3.2222222222222223,
-      "grad_norm": 0.573445737361908,
-      "kl": 0.0007366871383662025,
+      "completion_length": 563.1428571428571,
+      "epoch": 5.363636363636363,
+      "grad_norm": 0.5356454253196716,
+      "kl": 0.0006699909738797162,
       "learning_rate": 8.067960709356478e-08,
       "loss": 0.0,
-      "reward": 1.3570812046527863,
-      "reward_std": 1.19088014960289,
-      "rewards/concensus_correctness_reward_func": 0.025249999947845936,
-      "rewards/consensus_reward_func": 0.6666666666666666,
+      "reward": 0.7650677136012486,
+      "reward_std": 0.9540353034223829,
+      "rewards/concensus_correctness_reward_func": 0.0,
+      "rewards/consensus_reward_func": 0.07142857142857142,
       "rewards/cumulative_reward_2": 0.0,
       "rewards/final_correctness_reward_func": 0.0,
-      "rewards/question_recreation_reward_func": 0.6000395516554514,
+      "rewards/question_recreation_reward_func": 0.5341748382364001,
       "rewards/soft_format_reward_func": 0.0,
       "rewards/strict_format_reward_func": 0.0,
-      "rewards/xmlcount_reward_func": 0.06512501090765,
+      "rewards/xmlcount_reward_func": 0.15946428797074727,
       "step": 16
     },
     {
-      "completion_length": 391.9375,
-      "epoch": 3.6666666666666665,
-      "grad_norm": 0.5387791991233826,
-      "kl": 0.0007146395691961516,
+      "completion_length": 521.3928571428571,
+      "epoch": 6.0,
+      "grad_norm": 0.5471972227096558,
+      "kl": 0.0007351575976437223,
       "learning_rate": 3.013156219837776e-08,
       "loss": 0.0,
-      "reward": 1.735770734027028,
-      "reward_std": 1.017401821911335,
-      "rewards/concensus_correctness_reward_func": 0.11150000337511301,
-      "rewards/consensus_reward_func": 0.3125,
+      "reward": 1.1135373541287013,
+      "reward_std": 1.0160424539021082,
+      "rewards/concensus_correctness_reward_func": 0.04514285922050476,
+      "rewards/consensus_reward_func": 0.2857142857142857,
       "rewards/cumulative_reward_2": 0.0,
-      "rewards/final_correctness_reward_func": 0.25,
-      "rewards/question_recreation_reward_func": 0.5119269378483295,
+      "rewards/final_correctness_reward_func": 0.0,
+      "rewards/question_recreation_reward_func": 0.5741445156080383,
       "rewards/soft_format_reward_func": 0.0,
-      "rewards/strict_format_reward_func": 0.015625,
-      "rewards/xmlcount_reward_func": 0.5342187657952309,
+      "rewards/strict_format_reward_func": 0.03571428571428571,
+      "rewards/xmlcount_reward_func": 0.17282143022332871,
       "step": 18
     },
     {
-      "completion_length": 489.4583333333333,
-      "epoch": 4.0,
-      "grad_norm": 0.40115249156951904,
-      "kl": 0.0006926212323984752,
+      "completion_length": 502.125,
+      "epoch": 6.7272727272727275,
+      "grad_norm": 0.4341804087162018,
+      "kl": 0.0006616544378630351,
       "learning_rate": 3.4096741493194193e-09,
       "loss": 0.0,
-      "reward": 1.7193412780761719,
-      "reward_std": 1.9359967311223347,
-      "rewards/concensus_correctness_reward_func": 0.03125,
-      "rewards/consensus_reward_func": 0.9166666666666666,
+      "reward": 1.3489604517817497,
+      "reward_std": 0.9606379866600037,
+      "rewards/concensus_correctness_reward_func": 0.05925000458955765,
+      "rewards/consensus_reward_func": 0.5625,
       "rewards/cumulative_reward_2": 0.0,
-      "rewards/final_correctness_reward_func": 0.08333333333333333,
-      "rewards/question_recreation_reward_func": 0.5580913225809733,
-      "rewards/soft_format_reward_func": 0.0,
-      "rewards/strict_format_reward_func": 0.020833333333333332,
-      "rewards/xmlcount_reward_func": 0.10916665693124135,
+      "rewards/final_correctness_reward_func": 0.0,
+      "rewards/question_recreation_reward_func": 0.5194604564458132,
+      "rewards/soft_format_reward_func": 0.015625,
+      "rewards/strict_format_reward_func": 0.0,
+      "rewards/xmlcount_reward_func": 0.19212500751018524,
       "step": 20
     },
     {
-      "epoch": 4.0,
+      "epoch": 6.7272727272727275,
       "step": 20,
       "total_flos": 0.0,
-      "train_loss": 7.482587704998878e-07,
-      "train_runtime": 902.2542,
-      "train_samples_per_second": 0.355,
-      "train_steps_per_second": 0.022
+      "train_loss": 6.444601922339643e-07,
+      "train_runtime": 1608.3431,
+      "train_samples_per_second": 0.199,
+      "train_steps_per_second": 0.012
     }
   ],
   "logging_steps": 2,
   "max_steps": 20,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 5,
+  "num_train_epochs": 10,
   "save_steps": 25,
   "stateful_callbacks": {
     "TrainerControl": {