Beta
Explore
Marketplace
Neural Labs
Chat
Wallet
Docs
bidir-orm-dream7b-gsm8k – AI Model by AnonyRepo | AlphaNeural AI
You can deploy this model and start earning money today!
AnonyRepo
/
bidir-orm-dream7b-gsm8k
like
0
peft
outcome-reward-model
discrete-diffusion
gsm8k
lora
bidirectional
Dream-org/Dream-v0-Instruct-7B
adapter
mit
us
Views
No views yet
Model card
Files and Versions
Community
API
Deploy
Bidirectional Outcome Reward Model (Dream-7B, GSM8K)
LoRA adapter + reward head trained on final (mask=0) states for ORM reranking of Dream-7B Best-of-N samples. Used as
ORM-Rerank@N
in the paper's compute-matched Pareto.
Details
Base: Dream-org/Dream-v0-Instruct-7B (frozen)
Attention: bidirectional
Training data: final states only (mask=0)
No step embedding
Training: 8,407 steps (~6 epochs on 42,208 final-state samples), seed 42
Final-state accuracy at mask=0: 0.918 (s42/s43 mean)
