初始化项目，由ModelHub XC社区提供模型

Model: SeongryongJung/Qwen3-4B-Chemical-RLSD-TR
Source: Original Platform

results/hyperparameters.csv

@@ -0,0 +1,84 @@
+section,parameter,value,source
+Run identity,Base model,Qwen/Qwen3-4B,queue/script override
+Run identity,Dataset,Chemical / SciKnowEval chemistry,run_qwen3_generalization.sh
+Run identity,Method,RLSD_TR,run_qwen3_generalization.sh
+Run identity,Config,rlsd,run_qwen3_generalization.sh
+Run identity,Experiment,qwen3gen-chemistry-RLSD_TR-Qwen-Qwen3-4B-mbs8-decay0-tr0.1-train32-rollout8-lr1e-6-vllm0.8,run_qwen3_generalization.sh
+Run identity,W&B run,run-20260702_062054-bz6p2yxy,wandb
+Data,Train file,datasets/sciknoweval/chemistry/train.parquet,script override
+Data,Validation file,datasets/sciknoweval/chemistry/test.parquet,script override
+Data,Train batch size,32,queue/script override
+Data,Train max samples,3200,queue/script override
+Data,Prompt key,prompt,legacy_data.yaml default
+Data,Reward key,data_source,legacy_data.yaml default
+Data,Shuffle train data,True,user.yaml / legacy_data.yaml
+Data,Validation shuffle,False,legacy_data.yaml default
+Data,Filter overlong prompts,True,user.yaml
+Data,Prompt truncation,error,legacy_data.yaml default
+Data,enable_thinking,false,script override
+Schedule,Total training steps,100,queue/script override
+Schedule,Total epochs,30,ppo_trainer/user.yaml default
+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
+Sequence,Actor max token length per GPU,10240,queue/script override
+Rollout,Rollout engine,vllm,user.yaml
+Rollout,Rollout dtype,bfloat16,rollout.yaml default
+Rollout,Train rollout n,8,queue/script override
+Rollout,Train rollout temperature,1.0,script override
+Rollout,Train rollout top_p,1.0,script override
+Rollout,Train rollout do_sample,True,rollout.yaml default
+Rollout,Calculate rollout log probs,True,rlsd.yaml / script override
+Rollout,Max num batched tokens,10240,queue/script override
+Rollout,vLLM GPU memory utilization,0.8,queue/script override
+Rollout,Tensor model parallel size,2,rollout.yaml default
+Rollout,Free cache engine,True,rollout.yaml default
+Validation,Validation rollout n,16,queue/script override
+Validation,Validation temperature,0.6,queue/script override
+Validation,Validation top_p,0.95,queue/script override
+Validation,Validation do_sample,True,queue/script override
+Optimization,Optimizer,AdamW,fsdp optimizer config
+Optimization,Learning rate,1e-6,RLSD_TR method override
+Optimization,LR scheduler,constant,W&B config
+Optimization,LR warmup steps,10,script override
+Optimization,Weight decay,0.01,script override
+Optimization,Betas,"(0.9, 0.999)",W&B config
+Optimization,Gradient clip,1.0,script override
+PPO/GRPO,Advantage estimator,grpo,rlsd.yaml
+PPO/GRPO,Normalize GRPO advantages by std,False,script override
+PPO/GRPO,PPO epochs,1,W&B config
+PPO/GRPO,PPO mini batch size,8,queue/script override
+PPO/GRPO,PPO micro batch size per GPU,1,user.yaml
+PPO/GRPO,Clip ratio low,0.2,script override
+PPO/GRPO,Clip ratio high,0.28,script override
+PPO/GRPO,Gamma,1.0,ppo_trainer.yaml default
+PPO/GRPO,Lambda,1.0,ppo_trainer.yaml default
+PPO/GRPO,Use KL in reward,False,ppo_trainer/user.yaml
+PPO/GRPO,Actor KL loss observed,0.0,output.log
+Rollout correction,Importance sampling mode,token,script override
+Rollout correction,IS threshold,2.0,script override
+RLSD_TR,Policy loss mode,rlsd,method override
+RLSD_TR,Teacher regularization,trust-region,method override
+RLSD_TR,Trust-region mix / teacher update rate,0.1,queue/script override
+RLSD_TR,Token reweight lambda,0.5,queue/script override
+RLSD_TR,Token reweight eps_w,0.2,queue/script override
+RLSD_TR,Token reweight decay steps,0,queue/script override
+RLSD_TR,Max reprompt length,10240,method override
+RLSD_TR,Fused kernels,False,method override
+FSDP/System,Actor strategy,fsdp,dp_actor.yaml
+FSDP/System,FSDP dtype,bfloat16,W&B config
+FSDP/System,FSDP model dtype,fp32,W&B config
+FSDP/System,Use torch compile,True,W&B config
+FSDP/System,GPUs per node,8,queue/script override
+FSDP/System,Nodes,1,user.yaml
+FSDP/System,GPU type,NVIDIA H200,wandb-metadata
+Checkpoint/Logging,Checkpoint root,checkpoints/datasets/sciknoweval/chemistry,script override
+Checkpoint/Logging,Latest checkpointed iteration,100,latest_checkpointed_iteration.txt
+Checkpoint/Logging,Max actor checkpoints to keep,1,user.yaml
+Checkpoint/Logging,Logger,"console, wandb",ppo_trainer.yaml
+Checkpoint/Logging,W&B entity,seongryongjung-chung-ang-university,environment
+Checkpoint/Logging,W&B project,SDPO-root,user.yaml project_name
+Checkpoint/Logging,W&B group,QWEN3-RLSD-TR-GRPO-matched-generalization,method override

results/summary.json

@@ -0,0 +1,24 @@
+{
+  "repo_id": "SeongryongJung/Qwen3-4B-Chemical-RLSD-TR",
+  "output_dir": "/mnt/mole/SDPO/L2T/hf_upload_tr/Qwen3-4B-Chemical-RLSD-TR",
+  "best_step": 100,
+  "best_val_mean16": 0.6895833333333333,
+  "final_step": 100,
+  "final_val_mean16": 0.6895833333333333,
+  "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-00002.safetensors",
+    "model-00002-of-00002.safetensors"
+  ]
+}

results/training_score.png

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4fd721b8c29820fe41373b5c16ac993329102e09e022b7bd3a9f6f56b3ccee2a
+size 123640

results/training_scores.csv

@@ -0,0 +1,91 @@
+step,critic_score_mean
+1,0.5546875
+2,0.3828125
+3,0.453125
+4,0.33984375
+5,0.484375
+6,0.44140625
+7,0.40625
+8,0.52734375
+9,0.32421875
+11,0.421875
+12,0.4609375
+13,0.47265625
+14,0.453125
+15,0.4921875
+16,0.42578125
+17,0.41015625
+18,0.421875
+19,0.515625
+21,0.375
+22,0.375
+23,0.59375
+24,0.5078125
+25,0.48828125
+26,0.62890625
+27,0.5
+28,0.5859375
+29,0.34375
+31,0.625
+32,0.5625
+33,0.65234375
+34,0.515625
+35,0.5078125
+36,0.64453125
+37,0.515625
+38,0.59765625
+39,0.59765625
+41,0.59375
+42,0.5859375
+43,0.62890625
+44,0.6171875
+45,0.53515625
+46,0.54296875
+47,0.609375
+48,0.5859375
+49,0.65234375
+51,0.5390625
+52,0.59375
+53,0.57421875
+54,0.50390625
+55,0.609375
+56,0.64453125
+57,0.578125
+58,0.64453125
+59,0.58203125
+61,0.65234375
+62,0.64453125
+63,0.48828125
+64,0.70703125
+65,0.74609375
+66,0.546875
+67,0.66796875
+68,0.4765625
+69,0.56640625
+71,0.5859375
+72,0.71484375
+73,0.546875
+74,0.62109375
+75,0.73046875
+76,0.5625
+77,0.7265625
+78,0.65234375
+79,0.59765625
+81,0.640625
+82,0.625
+83,0.5625
+84,0.68359375
+85,0.69921875
+86,0.5625
+87,0.53125
+88,0.57421875
+89,0.625
+91,0.60546875
+92,0.625
+93,0.87890625
+94,0.73046875
+95,0.6640625
+96,0.6015625
+97,0.52734375
+98,0.58203125
+99,0.640625

results/validation_mean16.csv

@@ -0,0 +1,11 @@
+step,val_mean16
+10,0.4398809523809524
+20,0.5101190476190476
+30,0.5943452380952381
+40,0.6357142857142857
+50,0.6586309523809524
+60,0.6735119047619048
+70,0.6851190476190476
+80,0.6788690476190476
+90,0.6821428571428572
+100,0.6895833333333333