© 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. 75565
    Prev
    Next
    INDEX
    Explanations

    code structure and documentation

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

    The neuron is picking out running English words in prose (i.e. natural-language content words) and ignores most markup or punctuation.

    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
    application
    -0.72
    playback
    -0.71
    turno
    -0.70
    apper
    -0.69
    庑
    -0.67
    獾
    -0.66
    Szer
    -0.66
    Require
    -0.66
    ladı
    -0.66
    Hos
    -0.65
    POSITIVE LOGITS
    лку
    0.71
     Thors
    0.69
    访问
    0.67
    瞰
    0.66
    activ
    0.65
    бря
    0.63
    ffel
    0.62
    activa
    0.61
    onium
    0.61
    activation
    0.61
    Activations Density 0.070%

    No Known Activations