© 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. 26003
    Prev
    Next
    INDEX
    Explanations

    Copyright or legal statements

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

    This neuron fires on tokens that occur as the first meaningful element of a new block or heading line (e.g. section titles, comment markers, version labels, or the first verb in a definitional sentence).

    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
    wak
    -0.74
    š
    -0.71
    缸
    -0.69
    🔨
    -0.69
     Ini
    -0.68
    perso
    -0.66
    Lights
    -0.66
    '[
    -0.66
    chten
    -0.66
    Ini
    -0.66
    POSITIVE LOGITS
    allero
    0.64
    owanym
    0.64
    tax
    0.63
    ƒ
    0.63
     technique
    0.62
     gọi
    0.62
    dao
    0.60
    奪
    0.59
    æv
    0.59
    ľov
    0.59
    Activations Density 0.143%

    No Known Activations