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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 34280
    Prev
    Next
    INDEX
    Explanations

    abstract concepts and states

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

    The neuron fires on abstract, conceptual nouns (especially nominalizations like purpose, importance, bias, characteristics, feelings, attention, success).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
    دانلود
    -1.37
    場所は
    -1.33
    してます
    -1.29
    内容は
    -1.27
    写真は
    -1.17
    作り方
    -1.15
    してました
    -1.14
    呖
    -1.12
    院長
    -1.12
     作り方
    -1.12
    POSITIVE LOGITS
     of
    1.48
    あー
    1.28
    他又
    1.24
     waarschijnlijk
    1.23
    _
    1.22
     takich
    1.18
     verschillende
    1.17
    ほら
    1.15
    に入れる
    1.14
     niets
    1.13
    Activations Density 0.035%

    No Known Activations