© 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. 10-GEMMASCOPE-RES-131K
    7. 42014
    Prev
    Next
    INDEX
    Explanations

    Imp- prefix

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

    The neuron activates on any token containing the substring “imp,” i.e. words beginning with or including that “imp” root.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
    C
    -3.41
    D
    -3.38
    N
    -3.36
    L
    -3.17
    —-
    -2.66
    M
    -2.55
    A
    -2.50
    ăzi
    -2.42
     Erfolge
    -2.41
     idyllic
    -2.38
    POSITIVE LOGITS
    檍
    2.63
    you
    2.58
    婓
    2.52
    踅
    2.50
    鈇
    2.45
     klaus
    2.44
    橚
    2.44
    ſtra
    2.42
     drau
    2.39
    StreetMap
    2.39
    Activations Density 0.042%

    No Known Activations