Two LoRA cold-start SFT experiments teaching structured think/answer reasoning to Nanbeige4-3B-Base using distilled traces from frontier models
Mrinaal Arora
mrinaalarora
AI & ML interests
None yet
Recent Activity
updated a Space about 13 hours ago
mrinaalarora/textarena-wordle-env updated a Space about 17 hours ago
mrinaalarora/drylabsim published a Space 1 day ago
mrinaalarora/drylabsimOrganizations
None yet
spaces 4
pinned
Running
RL
DryLabSim
🚀
LLM agents plan noisy biological experiment pipelines
Running
RL
Textarena Environment Server
👀
Interacting with TextArena games via a web chat interface
Running
RL
Json Cleaning Env Environment Server
🖥
Clean messy JSON to fit a target schema
Sleeping
2
Qwen3-1.7B Wordle GRPO Training Dashboard
📈
Live training metrics for Qwen3-1.7B GRPO run on Wordle
models 4
mrinaalarora/wordle-grpo-Qwen3-1.7B
Reinforcement Learning • 2B • Updated • 932
mrinaalarora/Nanbeige4-3B-Cold-Start-Reasoning-LoRA-Opus-Epoch3
Text Generation • Updated • 33
mrinaalarora/nanbeige4-3b-cold-start-reasoning-lora-glm-12k
Text Generation • Updated • 30
mrinaalarora/Nanbeige4-3B-Cold-Start-Reasoning-LoRA
Text Generation • Updated • 2
datasets 0
None public yet