初始化项目，由ModelHub XC社区提供模型

Model: SeongryongJung/Qwen3-8B-Biology-GRPO-TR
Source: Original Platform

results/hyperparameters.csv

@@ -0,0 +1,33 @@
+section,parameter,value,source
+Run identity,Base model,Qwen/Qwen3-8B,queue/script override
+Run identity,Dataset,Biology / SciKnowEval biology,run_qwen3_generalization.sh
+Run identity,Method,GRPO,run_qwen3_generalization.sh
+Run identity,Config,baseline_grpo,run_qwen3_generalization.sh
+Run identity,Experiment,qwen3gen-biology-GRPO-Qwen-Qwen3-8B-mbs8-train32-rollout8-lr1e-6-vllm0.8,run_qwen3_generalization.sh
+Run identity,W&B run,run-20260703_072949-j4sb7trr,wandb
+Data,Train file,datasets/sciknoweval/biology/train.parquet,script override
+Data,Validation file,datasets/sciknoweval/biology/test.parquet,script override
+Data,Train batch size,32,queue/script override
+Data,Train max samples,3200,queue/script override
+Schedule,Total training steps,100,queue/script override
+Schedule,Validation before train,False,queue/script override
+Schedule,Save frequency,10,queue/script override
+Schedule,Validation frequency,10,queue/script override
+Sequence,Max prompt length,2048,queue/script override
+Sequence,Max response length,8192,queue/script override
+Sequence,Max model length,10240,queue/script override
+Rollout,Train rollout n,8,queue/script override
+Rollout,Validation rollout n,16,queue/script override
+Rollout,vLLM GPU memory utilization,0.8,queue/script override
+Optimization,Learning rate,1e-6,GRPO method override
+Optimization,Weight decay,0.01,script override
+PPO/GRPO,PPO mini batch size,8,queue/script override
+PPO/GRPO,Normalize GRPO advantages by std,False,baseline_grpo.yaml / script override
+Rollout correction,Importance sampling mode,token,script override
+Rollout correction,IS threshold,2.0,script override
+Checkpoint/Logging,Checkpoint root,checkpoints/datasets/sciknoweval/biology/qwen3gen-biology-GRPO-Qwen-Qwen3-8B-mbs8-train32-rollout8-lr1e-6-vllm0.8,script override
+Checkpoint/Logging,Latest checkpointed iteration,100,latest_checkpointed_iteration.txt
+Checkpoint/Logging,External actor archive,checkpoints/datasets/sciknoweval/biology/qwen3gen-biology-GRPO-Qwen-Qwen3-8B-mbs8-train32-rollout8-lr1e-6-vllm0.8/_actor_archive,preserve_actor_checkpoints.py
+Checkpoint/Logging,Logger,"console, wandb",ppo_trainer.yaml
+PPO/GRPO,Policy loss mode,vanilla,method override
+PPO/GRPO,Actor KL loss coef,0.0,method override

results/summary.json

@@ -0,0 +1,29 @@
+{
+  "repo_id": "SeongryongJung/Qwen3-8B-Biology-GRPO-TR",
+  "output_dir": "/mnt/mole/SDPO/L2T/hf_upload_tr/Qwen3-8B-Biology-GRPO-TR",
+  "experiment": "qwen3gen-biology-GRPO-Qwen-Qwen3-8B-mbs8-train32-rollout8-lr1e-6-vllm0.8",
+  "best_step": 90,
+  "best_val_mean16": 0.50375,
+  "best_actor_dir": "/mnt/mole/SDPO/L2T/checkpoints/datasets/sciknoweval/biology/qwen3gen-biology-GRPO-Qwen-Qwen3-8B-mbs8-train32-rollout8-lr1e-6-vllm0.8/global_step_90/actor",
+  "final_step": 100,
+  "final_val_mean16": 0.49875,
+  "final_actor_dir": "/mnt/mole/SDPO/L2T/checkpoints/datasets/sciknoweval/biology/qwen3gen-biology-GRPO-Qwen-Qwen3-8B-mbs8-train32-rollout8-lr1e-6-vllm0.8/global_step_100/actor",
+  "train_rows": 90,
+  "val_rows": 10,
+  "hf_model_files": [
+    "added_tokens.json",
+    "chat_template.jinja",
+    "config.json",
+    "generation_config.json",
+    "merges.txt",
+    "model.safetensors.index.json",
+    "special_tokens_map.json",
+    "tokenizer.json",
+    "tokenizer_config.json",
+    "vocab.json",
+    "model-00001-of-00004.safetensors",
+    "model-00002-of-00004.safetensors",
+    "model-00003-of-00004.safetensors",
+    "model-00004-of-00004.safetensors"
+  ]
+}

results/training_score.png

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1b568af928851953a7b6d7f121f8e8d2d71221c7bf70b4ead908f796959693f6
+size 140636

results/training_scores.csv

@@ -0,0 +1,91 @@
+step,critic_score_mean
+1,0.35546875
+2,0.41015625
+3,0.46875
+4,0.37109375
+5,0.484375
+6,0.39453125
+7,0.3359375
+8,0.28515625
+9,0.3359375
+11,0.375
+12,0.43359375
+13,0.3125
+14,0.35546875
+15,0.3671875
+16,0.453125
+17,0.23046875
+18,0.34765625
+19,0.453125
+21,0.4765625
+22,0.51171875
+23,0.3515625
+24,0.37890625
+25,0.40625
+26,0.37109375
+27,0.38671875
+28,0.3671875
+29,0.45703125
+31,0.35546875
+32,0.47265625
+33,0.4765625
+34,0.37890625
+35,0.50390625
+36,0.3828125
+37,0.4296875
+38,0.4375
+39,0.5546875
+41,0.50390625
+42,0.40625
+43,0.53515625
+44,0.47265625
+45,0.4375
+46,0.41796875
+47,0.390625
+48,0.5390625
+49,0.42578125
+51,0.4609375
+52,0.54296875
+53,0.4140625
+54,0.44140625
+55,0.546875
+56,0.4765625
+57,0.5546875
+58,0.46484375
+59,0.46875
+61,0.38671875
+62,0.5078125
+63,0.59765625
+64,0.5859375
+65,0.5546875
+66,0.46484375
+67,0.48046875
+68,0.62890625
+69,0.40625
+71,0.5234375
+72,0.4765625
+73,0.5703125
+74,0.44921875
+75,0.51953125
+76,0.3671875
+77,0.609375
+78,0.66015625
+79,0.4765625
+81,0.5625
+82,0.62109375
+83,0.54296875
+84,0.51171875
+85,0.45703125
+86,0.51953125
+87,0.484375
+88,0.453125
+89,0.6328125
+91,0.4765625
+92,0.6015625
+93,0.55859375
+94,0.44921875
+95,0.61328125
+96,0.59375
+97,0.62890625
+98,0.46484375
+99,0.49609375

results/validation_mean16.csv

@@ -0,0 +1,11 @@
+step,val_mean16
+10,0.305
+20,0.29875
+30,0.41125
+40,0.39125
+50,0.37625
+60,0.475
+70,0.46
+80,0.4875
+90,0.50375
+100,0.49875