{
"$type": "site.standard.document",
"bskyPostRef": {
"cid": "bafyreif3lsh7prov6fomn465uetvycdwrdpljfbxcd4cthx7xunmhcirmu",
"uri": "at://did:plc:pgryn3ephfd2xgft23qokfzt/app.bsky.feed.post/3mih2mnzan472"
},
"path": "/t/clanker-452kb-emotional-scoring-engine-strapped-to-llama-1b/174857#post_1",
"publishedAt": "2026-04-01T14:20:39.000Z",
"site": "https://discuss.huggingface.co",
"tags": [
"https://huggingface.co/spaces/deucebucket/clanker"
],
"textContent": "Built a deterministic engine that computes 7D emotional coordinates (VADUGWI) from text structure. Hooked it up to Llama-3.2-1B in a Gradio Space.\n\nThe model generates dialogue between two characters. The engine scores every line on 7 dimensions and tracks how each message shifts the other character’s emotional baseline. State carries forward through A+B=C transitions.\n\nWhat the engine does that the model can’t:\n\n * Detects 26 structural patterns (VICTIMIZATION, SELF_NULLIFY, SARCASM_INVERSION, CHOPPER_SPLIT, etc.)\n * Tracks self-worth (W) separately from valence – blaming yourself reads differently than blaming the world\n * Tracks intent direction (I) – reaching out vs pulling away vs commanding\n * Runs at 0.15ms/sentence on CPU, ~452KB total\n\n\n\nThe Space has two tabs:\n\n 1. Two characters argue (pick personalities, give them a topic, watch the emotional states shift)\n 2. Score any text (type anything, see the raw coordinates)\n\n\n\nWhat I’m working on now: training a Phi-4-mini LoRA to actually learn the VADUGWI math so it can compute emotional coordinates internally, not just generate text.\n\nSpace: https://huggingface.co/spaces/deucebucket/clanker\n\nWould love feedback on what the engine gets right and where it falls apart. Try typing something sarcastic in the Score tab.",
"title": "Clanker – 452KB emotional scoring engine strapped to Llama-1B"
}