© 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
    2. Gemma-2-27B
    3. 34-GEMMASCOPE-RES-131K
    4. 90685
    Prev
    Next
    INDEX
    Explanations

    comment code structure

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

    The neuron responds to the most frequently repeated tokens in the snippet (e.g. “in”, “help”, “%%”), i.e. it’s a high‐frequency token detector.

    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
     Tagged
    -0.79
    lyt
    -0.73
    elry
    -0.72
    acus
    -0.72
    ള്ള
    -0.70
     receber
    -0.69
    orship
    -0.69
    妹
    -0.68
    indx
    -0.68
     facilitar
    -0.68
    POSITIVE LOGITS
    ombies
    0.84
     н
    0.77
     Пред
    0.75
    ysław
    0.74
    ラミ
    0.72
    skab
    0.72
    скра
    0.72
    cze
    0.71
     pomo
    0.71
     Maloney
    0.71
    Activations Density 0.004%

    No Known Activations