© 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. 102480
    Prev
    Next
    INDEX
    Explanations

    blockbusters, Foundation, recorded, license

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

    The neuron primarily fires on numeric tokens or number-like substrings (e.g. standalone digits, years, version numbers).

    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
     Scalars
    -0.79
    Authorities
    -0.76
     Hodge
    -0.74
    alsch
    -0.74
    fstream
    -0.74
    隈
    -0.72
    BTC
    -0.69
     Orthop
    -0.69
    つけ
    -0.69
     quê
    -0.69
    POSITIVE LOGITS
    Convergence
    0.72
     stitches
    0.65
    anja
    0.64
    AppSettings
    0.63
     shouldn
    0.63
    aksanaan
    0.61
     jajaran
    0.61
    mpto
    0.61
    メン
    0.60
    sodik
    0.60
    Activations Density 0.061%

    No Known Activations