初始化项目，由ModelHub XC社区提供模型

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

results/export_note.csv

@@ -0,0 +1,8 @@
+key,value
+root_checkpoint_step,90
+root_checkpoint_kind,best_validation
+root_actor_dir,checkpoints/datasets/sciknoweval/biology/qwen3gen-biology-RLSD_TR-Qwen-Qwen3-4B-mbs8-decay0-tr0.1-train32-rollout8-lr1e-6-vllm0.8/_actor_archive/global_step_90/actor
+last_checkpoint_step,100
+last_actor_dir,checkpoints/datasets/sciknoweval/biology/qwen3gen-biology-RLSD_TR-Qwen-Qwen3-4B-mbs8-decay0-tr0.1-train32-rollout8-lr1e-6-vllm0.8/global_step_100/actor
+best_validation_mean16,0.485000000000
+final_validation_mean16,0.428750000000

results/hyperparameters.csv

@@ -0,0 +1,85 @@
+section,parameter,value,source
+Run identity,Base model,Qwen/Qwen3-4B,queue/script override
+Run identity,Dataset,Biology / SciKnowEval biology,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-biology-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_113907-xauqhwk1,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
+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/biology,script override
+Checkpoint/Logging,Latest checkpointed iteration,100,latest_checkpointed_iteration.txt
+Checkpoint/Logging,Max actor checkpoints to keep,1 + external _actor_archive preservation,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
+Checkpoint/Logging,External actor archive,checkpoints/datasets/sciknoweval/biology/qwen3gen-biology-RLSD_TR-Qwen-Qwen3-4B-mbs8-decay0-tr0.1-train32-rollout8-lr1e-6-vllm0.8/_actor_archive,preserve_actor_checkpoints.py

results/summary.json

@@ -0,0 +1,42 @@
+{
+  "repo_id": "SeongryongJung/Qwen3-4B-Biology-RLSD-TR",
+  "output_dir": "/mnt/mole/SDPO/L2T/hf_upload_tr/Qwen3-4B-Biology-RLSD-TR",
+  "root_checkpoint_step": 90,
+  "root_checkpoint_kind": "best_validation",
+  "best_step": 90,
+  "best_val_mean16": 0.485,
+  "best_actor_dir": "/mnt/mole/SDPO/L2T/checkpoints/datasets/sciknoweval/biology/qwen3gen-biology-RLSD_TR-Qwen-Qwen3-4B-mbs8-decay0-tr0.1-train32-rollout8-lr1e-6-vllm0.8/_actor_archive/global_step_90/actor",
+  "final_step": 100,
+  "final_val_mean16": 0.42875,
+  "final_actor_dir": "/mnt/mole/SDPO/L2T/checkpoints/datasets/sciknoweval/biology/qwen3gen-biology-RLSD_TR-Qwen-Qwen3-4B-mbs8-decay0-tr0.1-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-00002.safetensors",
+    "model-00002-of-00002.safetensors"
+  ],
+  "hf_last_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:d8d179d4cbb80eb70d11587378ff10ca730624560a552bf74f79c6533df4aed7
+size 134385

results/training_scores.csv

@@ -0,0 +1,91 @@
+step,critic_score_mean
+1,0.3125
+2,0.4375
+3,0.34765625
+4,0.41796875
+5,0.47265625
+6,0.3125
+7,0.22265625
+8,0.28515625
+9,0.3984375
+11,0.421875
+12,0.421875
+13,0.34765625
+14,0.265625
+15,0.37890625
+16,0.48046875
+17,0.32421875
+18,0.359375
+19,0.328125
+21,0.390625
+22,0.46484375
+23,0.265625
+24,0.41015625
+25,0.40234375
+26,0.41015625
+27,0.3984375
+28,0.35546875
+29,0.4921875
+31,0.3828125
+32,0.4765625
+33,0.484375
+34,0.37890625
+35,0.4921875
+36,0.33203125
+37,0.3828125
+38,0.50390625
+39,0.4140625
+41,0.54296875
+42,0.3125
+43,0.4921875
+44,0.52734375
+45,0.4140625
+46,0.5390625
+47,0.484375
+48,0.46484375
+49,0.43359375
+51,0.37109375
+52,0.4609375
+53,0.4453125
+54,0.51171875
+55,0.5546875
+56,0.51953125
+57,0.60546875
+58,0.5390625
+59,0.53515625
+61,0.4296875
+62,0.51171875
+63,0.60546875
+64,0.4765625
+65,0.5234375
+66,0.53515625
+67,0.43359375
+68,0.52734375
+69,0.46484375
+71,0.48046875
+72,0.421875
+73,0.6171875
+74,0.50390625
+75,0.484375
+76,0.46484375
+77,0.63671875
+78,0.58984375
+79,0.50390625
+81,0.59765625
+82,0.59375
+83,0.5859375
+84,0.48046875
+85,0.48828125
+86,0.56640625
+87,0.42578125
+88,0.4921875
+89,0.66796875
+91,0.578125
+92,0.515625
+93,0.55078125
+94,0.515625
+95,0.60546875
+96,0.53515625
+97,0.515625
+98,0.47265625
+99,0.5859375

results/validation_mean16.csv

@@ -0,0 +1,11 @@
+step,val_mean16
+10,0.325
+20,0.32875
+30,0.3625
+40,0.37125
+50,0.41125
+60,0.40625
+70,0.42
+80,0.41875
+90,0.485
+100,0.42875