© 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. 109180
    Prev
    Next
    INDEX
    Explanations

    said followed by a name

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

    The neuron fires strongly on numeric and punctuation‐style tokens (e.g. dates, measurement figures, commas, apostrophes, etc.).

    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
    Ａ
    -0.92
    қы
    -0.89
    💺
    -0.89
    اقت
    -0.84
     Во
    -0.84
     Merge
    -0.83
    setScale
    -0.82
    PROXY
    -0.82
    tighet
    -0.81
    innerHeight
    -0.81
    POSITIVE LOGITS
     director
    1.44
     said
    1.40
     vice
    1.32
     Director
    1.23
     vp
    1.23
     deputy
    1.18
     VP
    1.18
     senior
    1.17
     Senior
    1.17
    SENIOR
    1.15
    Activations Density 0.035%

    No Known Activations