End of training

README.md

@@ -1,5 +1,5 @@
 ---
-base_model: Gensyn/Qwen2.5-0.5B-Instruct
+base_model: Gensyn/Qwen2.5-1.5B-Instruct
 library_name: transformers
 model_name: gensyn-checkpoints-meek_regal_badger
 tags:
@@ -8,13 +8,14 @@ tags:
 - grpo
 - gensyn
 - I am meek regal badger
+- unsloth
 - trl
 licence: license
 ---
 
 # Model Card for gensyn-checkpoints-meek_regal_badger
 
-This model is a fine-tuned version of [Gensyn/Qwen2.5-0.5B-Instruct](https://huggingface.co/Gensyn/Qwen2.5-0.5B-Instruct).
+This model is a fine-tuned version of [Gensyn/Qwen2.5-1.5B-Instruct](https://huggingface.co/Gensyn/Qwen2.5-1.5B-Instruct).
 It has been trained using [TRL](https://github.com/huggingface/trl).
 
 ## Quick start
@@ -39,7 +40,7 @@ This model was trained with GRPO, a method introduced in [DeepSeekMath: Pushing
 
 - TRL: 0.15.2
 - Transformers: 4.51.3
-- Pytorch: 2.5.1
+- Pytorch: 2.6.0
 - Datasets: 3.5.0
 - Tokenizers: 0.21.1
 

adapter_config.json

@@ -0,0 +1,39 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "Gensyn/Qwen2.5-1.5B-Instruct",
+  "bias": "none",
+  "corda_config": null,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 16,
+  "lora_bias": false,
+  "lora_dropout": 0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 16,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "k_proj",
+    "q_proj",
+    "gate_proj",
+    "up_proj",
+    "v_proj",
+    "down_proj",
+    "o_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_dora": false,
+  "use_rslora": false
+}

adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fb9dadfbe441e35e95c36071f0658550dcdba6a69a89bca55b14d5327b2cb1d1
+size 73911112

all_results.json

@@ -1,8 +1,8 @@
 {
     "total_flos": 0.0,
-    "train_loss": 9.198513089359039e-06,
-    "train_runtime": 215.0623,
-    "train_samples": 28,
-    "train_samples_per_second": 1.488,
-    "train_steps_per_second": 0.093
+    "train_loss": 7.838596616238647e-07,
+    "train_runtime": 1722.3344,
+    "train_samples": 11,
+    "train_samples_per_second": 0.186,
+    "train_steps_per_second": 0.012
 }

train_results.json

@@ -1,8 +1,8 @@
 {
     "total_flos": 0.0,
-    "train_loss": 9.198513089359039e-06,
-    "train_runtime": 215.0623,
-    "train_samples": 28,
-    "train_samples_per_second": 1.488,
-    "train_steps_per_second": 0.093
+    "train_loss": 7.838596616238647e-07,
+    "train_runtime": 1722.3344,
+    "train_samples": 11,
+    "train_samples_per_second": 0.186,
+    "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": 5.0,
+  "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": 298.0625,
-      "epoch": 0.5714285714285714,
-      "grad_norm": 6.717251777648926,
-      "kl": 0.0,
+      "completion_length": 510.15625,
+      "epoch": 0.7272727272727273,
+      "grad_norm": 13.140520095825195,
+      "kl": 0.0008186797858797945,
       "learning_rate": 5e-07,
       "loss": 0.0,
-      "reward": 2.065369065850973,
-      "reward_std": 1.6619373487774283,
-      "rewards/concensus_correctness_reward_func": 0.39149999618530273,
-      "rewards/consensus_reward_func": 0.5,
+      "reward": 1.2687874287366867,
+      "reward_std": 1.0542942136526108,
+      "rewards/concensus_correctness_reward_func": 0.12606249749660492,
+      "rewards/consensus_reward_func": 0.3125,
       "rewards/cumulative_reward_2": 0.0,
       "rewards/final_correctness_reward_func": 0.0,
-      "rewards/question_recreation_reward_func": 0.45577534078620374,
-      "rewards/soft_format_reward_func": 0.0,
-      "rewards/strict_format_reward_func": 0.03125,
-      "rewards/xmlcount_reward_func": 0.6868437463417649,
+      "rewards/question_recreation_reward_func": 0.6433187238872051,
+      "rewards/soft_format_reward_func": 0.015625,
+      "rewards/strict_format_reward_func": 0.015625,
+      "rewards/xmlcount_reward_func": 0.15565624134615064,
       "step": 2
     },
     {
-      "completion_length": 266.2916666666667,
-      "epoch": 1.0,
-      "grad_norm": 3.6867785453796387,
-      "kl": 0.001406152257307743,
+      "completion_length": 467.85714285714283,
+      "epoch": 1.3636363636363638,
+      "grad_norm": 0.3759777843952179,
+      "kl": 0.000642291282669508,
       "learning_rate": 4.864543104251586e-07,
       "loss": 0.0,
-      "reward": 2.4803318232297897,
-      "reward_std": 1.9100091128299634,
-      "rewards/concensus_correctness_reward_func": 0.5114166711767515,
-      "rewards/consensus_reward_func": 0.4166666666666667,
+      "reward": 1.3871630813394273,
+      "reward_std": 1.1937960088253021,
+      "rewards/concensus_correctness_reward_func": 0.04971428428377424,
+      "rewards/consensus_reward_func": 0.35714285714285715,
       "rewards/cumulative_reward_2": 0.0,
-      "rewards/final_correctness_reward_func": 0.25,
-      "rewards/question_recreation_reward_func": 0.49745685296754044,
+      "rewards/final_correctness_reward_func": 0.0,
+      "rewards/question_recreation_reward_func": 0.6371630557945797,
       "rewards/soft_format_reward_func": 0.0,
-      "rewards/strict_format_reward_func": 0.0,
-      "rewards/xmlcount_reward_func": 0.8047916715343794,
+      "rewards/strict_format_reward_func": 0.017857142857142856,
+      "rewards/xmlcount_reward_func": 0.3252857114587511,
       "step": 4
     },
     {
-      "completion_length": 204.0625,
-      "epoch": 1.5714285714285714,
-      "grad_norm": 8.32531452178955,
-      "kl": 0.011501419357955456,
+      "completion_length": 465.75,
+      "epoch": 2.0,
+      "grad_norm": 0.39969348907470703,
+      "kl": 0.0006992748531047255,
       "learning_rate": 4.472851273490984e-07,
       "loss": 0.0,
-      "reward": 2.601933002471924,
-      "reward_std": 0.9620060075540096,
-      "rewards/concensus_correctness_reward_func": 0.4566875025629997,
-      "rewards/consensus_reward_func": 0.4375,
+      "reward": 1.1130137613841466,
+      "reward_std": 0.9021646614585604,
+      "rewards/concensus_correctness_reward_func": 0.0,
+      "rewards/consensus_reward_func": 0.21428571428571427,
       "rewards/cumulative_reward_2": 0.0,
-      "rewards/final_correctness_reward_func": 0.0625,
-      "rewards/question_recreation_reward_func": 0.5837454923894256,
+      "rewards/final_correctness_reward_func": 0.0,
+      "rewards/question_recreation_reward_func": 0.6723351946898869,
       "rewards/soft_format_reward_func": 0.0,
-      "rewards/strict_format_reward_func": 0.09375,
-      "rewards/xmlcount_reward_func": 0.9677500203251839,
+      "rewards/strict_format_reward_func": 0.017857142857142856,
+      "rewards/xmlcount_reward_func": 0.20853571806635177,
       "step": 6
     },
     {
-      "completion_length": 227.25,
-      "epoch": 2.0,
-      "grad_norm": 4.771548271179199,
-      "kl": 0.006304021313553676,
+      "completion_length": 487.5625,
+      "epoch": 2.7272727272727275,
+      "grad_norm": 0.5062351822853088,
+      "kl": 0.001850860000558896,
       "learning_rate": 3.867370395306068e-07,
       "loss": 0.0,
-      "reward": 3.030232419570287,
-      "reward_std": 1.8522576310982306,
-      "rewards/concensus_correctness_reward_func": 0.7245000004768372,
-      "rewards/consensus_reward_func": 0.75,
+      "reward": 0.6859643496572971,
+      "reward_std": 0.946899801492691,
+      "rewards/concensus_correctness_reward_func": 0.040312498807907104,
+      "rewards/consensus_reward_func": 0.0625,
       "rewards/cumulative_reward_2": 0.0,
       "rewards/final_correctness_reward_func": 0.0,
-      "rewards/question_recreation_reward_func": 0.6421075016260147,
+      "rewards/question_recreation_reward_func": 0.5663705877959728,
       "rewards/soft_format_reward_func": 0.0,
-      "rewards/strict_format_reward_func": 0.020833333333333332,
-      "rewards/xmlcount_reward_func": 0.892791673541069,
+      "rewards/strict_format_reward_func": 0.0,
+      "rewards/xmlcount_reward_func": 0.01678125560283661,
       "step": 8
     },
     {
-      "completion_length": 249.625,
-      "epoch": 2.571428571428571,
-      "grad_norm": 6.5946149826049805,
-      "kl": 0.011220796084671747,
+      "completion_length": 503.57142857142856,
+      "epoch": 3.3636363636363638,
+      "grad_norm": 0.5845171809196472,
+      "kl": 0.0006387338674228106,
       "learning_rate": 3.1137137178519977e-07,
       "loss": 0.0,
-      "reward": 2.330574309453368,
-      "reward_std": 0.9070390482665971,
-      "rewards/concensus_correctness_reward_func": 0.3594375029206276,
-      "rewards/consensus_reward_func": 0.3125,
+      "reward": 1.3737128632409232,
+      "reward_std": 1.0307631875787462,
+      "rewards/concensus_correctness_reward_func": 0.03571428571428571,
+      "rewards/consensus_reward_func": 0.14285714285714285,
       "rewards/cumulative_reward_2": 0.0,
-      "rewards/final_correctness_reward_func": 0.1875,
-      "rewards/question_recreation_reward_func": 0.42963684926507995,
+      "rewards/final_correctness_reward_func": 0.07142857142857142,
+      "rewards/question_recreation_reward_func": 0.7071057600634438,
       "rewards/soft_format_reward_func": 0.0,
-      "rewards/strict_format_reward_func": 0.046875,
-      "rewards/xmlcount_reward_func": 0.9946250058710575,
+      "rewards/strict_format_reward_func": 0.0,
+      "rewards/xmlcount_reward_func": 0.4166071436234883,
       "step": 10
     },
     {
-      "completion_length": 284.0833333333333,
-      "epoch": 3.0,
-      "grad_norm": 8.51677131652832,
-      "kl": 0.01575400565828507,
+      "completion_length": 500.17857142857144,
+      "epoch": 4.0,
+      "grad_norm": 0.34793826937675476,
+      "kl": 0.0006221807769699288,
       "learning_rate": 2.2935516363191693e-07,
       "loss": 0.0,
-      "reward": 2.885760868589083,
-      "reward_std": 1.9988142798344295,
-      "rewards/concensus_correctness_reward_func": 0.710416667163372,
-      "rewards/consensus_reward_func": 0.75,
+      "reward": 0.890633191381182,
+      "reward_std": 1.0413332028048379,
+      "rewards/concensus_correctness_reward_func": 0.0,
+      "rewards/consensus_reward_func": 0.2857142857142857,
       "rewards/cumulative_reward_2": 0.0,
       "rewards/final_correctness_reward_func": 0.0,
-      "rewards/question_recreation_reward_func": 0.6493442542850971,
+      "rewards/question_recreation_reward_func": 0.5545260693345752,
       "rewards/soft_format_reward_func": 0.0,
-      "rewards/strict_format_reward_func": 0.041666666666666664,
-      "rewards/xmlcount_reward_func": 0.7343333313862482,
+      "rewards/strict_format_reward_func": 0.0,
+      "rewards/xmlcount_reward_func": 0.05039285868406296,
       "step": 12
     },
     {
-      "completion_length": 287.21875,
-      "epoch": 3.571428571428571,
-      "grad_norm": 5.919245719909668,
-      "kl": 0.01331145316362381,
+      "completion_length": 439.3125,
+      "epoch": 4.7272727272727275,
+      "grad_norm": 0.573300838470459,
+      "kl": 0.000715912836312782,
       "learning_rate": 1.4957614383675767e-07,
       "loss": 0.0,
-      "reward": 2.9632712304592133,
-      "reward_std": 1.4405226800590754,
-      "rewards/concensus_correctness_reward_func": 0.756249999627471,
-      "rewards/consensus_reward_func": 0.75,
+      "reward": 0.9845727905631065,
+      "reward_std": 1.0894903242588043,
+      "rewards/concensus_correctness_reward_func": 0.04349999874830246,
+      "rewards/consensus_reward_func": 0.1875,
       "rewards/cumulative_reward_2": 0.0,
       "rewards/final_correctness_reward_func": 0.0,
-      "rewards/question_recreation_reward_func": 0.6578337242826819,
+      "rewards/question_recreation_reward_func": 0.5931040458381176,
       "rewards/soft_format_reward_func": 0.0,
       "rewards/strict_format_reward_func": 0.015625,
-      "rewards/xmlcount_reward_func": 0.7835625000298023,
+      "rewards/xmlcount_reward_func": 0.14484376274049282,
       "step": 14
     },
     {
-      "completion_length": 290.2083333333333,
-      "epoch": 4.0,
-      "grad_norm": 5.046748161315918,
-      "kl": 0.023930478307496134,
+      "completion_length": 550.5,
+      "epoch": 5.363636363636363,
+      "grad_norm": 8.911335945129395,
+      "kl": 0.0011261084873694927,
       "learning_rate": 8.067960709356478e-08,
       "loss": 0.0,
-      "reward": 2.493785634636879,
-      "reward_std": 0.5378106032779518,
-      "rewards/concensus_correctness_reward_func": 0.34049999713897705,
-      "rewards/consensus_reward_func": 0.4166666666666667,
+      "reward": 1.0786219707557134,
+      "reward_std": 1.1641690645899092,
+      "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": 0.5334939612851789,
+      "rewards/final_correctness_reward_func": 0.0,
+      "rewards/question_recreation_reward_func": 0.5767291230814797,
       "rewards/soft_format_reward_func": 0.0,
-      "rewards/strict_format_reward_func": 0.041666666666666664,
-      "rewards/xmlcount_reward_func": 0.9114583308498064,
+      "rewards/strict_format_reward_func": 0.0,
+      "rewards/xmlcount_reward_func": 0.07332143187522888,
       "step": 16
     },
     {
-      "completion_length": 294.375,
-      "epoch": 4.571428571428571,
-      "grad_norm": 7.338254928588867,
-      "kl": 0.016075315224952647,
+      "completion_length": 489.2857142857143,
+      "epoch": 6.0,
+      "grad_norm": 0.37421098351478577,
+      "kl": 0.0005628815935259419,
       "learning_rate": 3.013156219837776e-08,
       "loss": 0.0,
-      "reward": 2.63186077773571,
-      "reward_std": 0.7248760172806215,
-      "rewards/concensus_correctness_reward_func": 0.4661874994635582,
-      "rewards/consensus_reward_func": 0.5,
+      "reward": 1.2610549884183067,
+      "reward_std": 0.7215567742075238,
+      "rewards/concensus_correctness_reward_func": 0.03571428571428571,
+      "rewards/consensus_reward_func": 0.14285714285714285,
       "rewards/cumulative_reward_2": 0.0,
-      "rewards/final_correctness_reward_func": 0.125,
-      "rewards/question_recreation_reward_func": 0.6226419649028685,
+      "rewards/final_correctness_reward_func": 0.0,
+      "rewards/question_recreation_reward_func": 0.6783763894012996,
       "rewards/soft_format_reward_func": 0.0,
-      "rewards/strict_format_reward_func": 0.046875,
-      "rewards/xmlcount_reward_func": 0.8711562715470791,
+      "rewards/strict_format_reward_func": 0.0,
+      "rewards/xmlcount_reward_func": 0.4041071449007307,
       "step": 18
     },
     {
-      "completion_length": 286.5416666666667,
-      "epoch": 5.0,
-      "grad_norm": 5.04390811920166,
-      "kl": 0.017868785148796935,
+      "completion_length": 498.3125,
+      "epoch": 6.7272727272727275,
+      "grad_norm": 0.35772696137428284,
+      "kl": 0.0007357651702477597,
       "learning_rate": 3.4096741493194193e-09,
       "loss": 0.0,
-      "reward": 2.4256584346294403,
-      "reward_std": 1.272670385427773,
-      "rewards/concensus_correctness_reward_func": 0.4596666644016902,
-      "rewards/consensus_reward_func": 0.4166666666666667,
+      "reward": 1.0650005312636495,
+      "reward_std": 0.9317956492304802,
+      "rewards/concensus_correctness_reward_func": 0.04349999874830246,
+      "rewards/consensus_reward_func": 0.125,
       "rewards/cumulative_reward_2": 0.0,
-      "rewards/final_correctness_reward_func": 0.08333333333333333,
-      "rewards/question_recreation_reward_func": 0.5910334512591362,
+      "rewards/final_correctness_reward_func": 0.0,
+      "rewards/question_recreation_reward_func": 0.6993755288422108,
       "rewards/soft_format_reward_func": 0.0,
-      "rewards/strict_format_reward_func": 0.0,
-      "rewards/xmlcount_reward_func": 0.8749583264191946,
+      "rewards/strict_format_reward_func": 0.015625,
+      "rewards/xmlcount_reward_func": 0.1815000008791685,
       "step": 20
     },
     {
-      "epoch": 5.0,
+      "epoch": 6.7272727272727275,
       "step": 20,
       "total_flos": 0.0,
-      "train_loss": 9.198513089359039e-06,
-      "train_runtime": 215.0623,
-      "train_samples_per_second": 1.488,
-      "train_steps_per_second": 0.093
+      "train_loss": 7.838596616238647e-07,
+      "train_runtime": 1722.3344,
+      "train_samples_per_second": 0.186,
+      "train_steps_per_second": 0.012
     }
   ],
   "logging_steps": 2,
   "max_steps": 20,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 7,
+  "num_train_epochs": 10,
   "save_steps": 25,
   "stateful_callbacks": {
     "TrainerControl": {
@@ -227,7 +227,7 @@
     }
   },
   "total_flos": 0.0,
-  "train_batch_size": 2,
+  "train_batch_size": 4,
   "trial_name": null,
   "trial_params": null
 }

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:47a25bb360e4b1beaddeb3ce0035c659a970fc0c77eae513be0ae042337a8e09
-size 5880
+oid sha256:bb851532fefc42dd7404541195a8d4a37b0b8a49334827f753cb9db442d454d7
+size 5944