{
  "$type": "site.standard.document",
  "bskyPostRef": {
    "cid": "bafyreid7fykicuqym762cmsh2t4gqhrjv22a2yviyg6dc4jitc2ibe742u",
    "uri": "at://did:plc:pgryn3ephfd2xgft23qokfzt/app.bsky.feed.post/3mj4ygksw52i2"
  },
  "path": "/t/space-lora-lens-behavior-studio-real-time-personality-steering-with-a-slider/175145#post_1",
  "publishedAt": "2026-04-10T09:04:05.000Z",
  "site": "https://discuss.huggingface.co",
  "tags": [
    "https://huggingface.co/spaces/Intuitivation/LoRALens-Demo"
  ],
  "textContent": "Hey everyone — I built a Space that takes a different approach to local LLM customization and wanted to share it here.\n\n**What it does:** Instead of tweaking system prompts to get a “more creative” or “more analytical” model, LoRA Lens Behavior Studio applies behavior-level adapters extracted via contrast pairs and SVD decomposition — then lets you dial the intensity up or down with a signed alpha slider in real time.\n\nThe core pipeline:\n\n  * Contrast pair extraction (target behavior vs. baseline)\n\n  * Delta compression into LoRA adapters via SVD (rank 32–64)\n\n  * Live `AlphaController` for non-destructive steering — nothing gets baked permanently\n\n\n\n\n**Demo:** https://huggingface.co/spaces/Intuitivation/LoRALens-Demo\n\nCertified across Mistral-7B, Llama-3-8B, and Qwen2.5 series. 36 behavior packs across Communication, Reasoning, Creative, Personality, Domain-Specific, and Experimental categories.\n\nThe thing that surprised people most in testing: dragging the slider from −1.0 to +1.0 on something like the Confident Clarity pack produces a noticeably different _feel_ in the outputs — not just word choice, but sentence structure and hedging behavior. It’s not a prompt effect.\n\nWould love feedback from anyone who’s worked with activation-level steering or has thoughts on the SVD rank tradeoffs. Happy to answer questions about the architecture.",
  "title": "[Space] LoRA Lens Behavior Studio — Real-time personality steering with a slider"
}