© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 670
    Prev
    Next
    INDEX
    Explanations

    preventing drafts and elements

    np_acts-logits-general · gemini-2.5-flash-lite

    The neuron fires on technical engineering terms related to sealing or weatherproofing window and door components.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     mecánico
    -0.91
     constitutional
    -0.83
     cachorros
    -0.80
     déb
    -0.79
     ostro
    -0.77
    𝑵
    -0.74
     provecho
    -0.74
    Sectors
    -0.74
     mecánica
    -0.73
    压制
    -0.73
    POSITIVE LOGITS
     draft
    1.63
     drafts
    1.62
    draft
    1.54
     seal
    1.53
     draught
    1.48
     sealing
    1.46
     Draft
    1.46
     seals
    1.43
    Draft
    1.38
    seal
    1.34
    Activations Density 0.016%

    No Known Activations