© 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. 59867
    Prev
    Next
    INDEX
    Explanations

    how to + action

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

    This neuron detects question‐style, how/where instructional phrases (e.g. “How to…,” “Where can I…,” “Where to…”) in headings or titles.

    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
    ssss
    -0.84
    czny
    -0.82
    quing
    -0.78
    าสตร์
    -0.77
    stmt
    -0.77
     REFER
    -0.76
    hebung
    -0.76
    汐
    -0.75
     Wik
    -0.73
    ⟡
    -0.73
    POSITIVE LOGITS
    roughts
    0.80
     veloce
    0.79
     lower
    0.75
    JOUR
    0.73
     wobei
    0.73
     jedną
    0.72
     superiore
    0.70
    ありますか
    0.69
    }".
    0.69
     différentes
    0.69
    Activations Density 0.015%

    No Known Activations