© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 14797
    Prev
    Next
    INDEX
    Explanations

    core programming constructs

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

    The neuron flags the single most salient domain‐specific keyword in each snippet—i.e. the main content word (often a technical or topic term) around which the text revolves.

    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
     prematurely
    -0.85
     або
    -0.83
    Sts
    -0.82
    Procedimento
    -0.81
     conectado
    -0.81
    tift
    -0.81
    ").
    -0.81
     ſeveral
    -0.80
     longterm
    -0.80
    主役
    -0.80
    POSITIVE LOGITS
    She
    0.93
    وير
    0.92
    ває
    0.88
     夜
    0.88
    Hospital
    0.84
    Her
    0.82
    ToAdd
    0.81
     säs
    0.81
    liche
    0.81
    ����
    0.80
    Activations Density 0.044%

    No Known Activations