© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 863
    Prev
    Next
    INDEX
    Explanations

    access and refresh tokens

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

    The neuron flags tokens made up of digits (numeric literals), outputting an activation equal to the number they represent.

    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
    Neutral
    -0.85
    neutral
    -0.79
    illotine
    -0.76
    𝗣
    -0.75
    feb
    -0.73
    roja
    -0.73
     Neutral
    -0.72
    hep
    -0.72
     indirizzo
    -0.71
    وست
    -0.71
    POSITIVE LOGITS
     Refresh
    1.73
     refresh
    1.59
    refresh
    1.42
    Refresh
    1.28
     refreshments
    1.25
    TokenType
    1.23
     refreshment
    1.20
    REFRESH
    1.16
    RefreshToken
    1.05
    refreshToken
    1.01
    Activations Density 0.011%

    No Known Activations