© 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. 10-GEMMASCOPE-RES-131K
    7. 70148
    Prev
    Next
    INDEX
    Explanations

    searching specific indices

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

    The neuron activates on numeric literals (any standalone numbers) within the text.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
    étit
    -1.91
     apparaat
    -1.88
     帅气
    -1.75
    驁
    -1.73
    涐
    -1.72
    椟
    -1.66
     namorados
    -1.66
    🏓
    -1.63
     搞笑
    -1.62
     aantal
    -1.60
    POSITIVE LOGITS
    1
    2.30
    4
    2.19
    of
    2.16
    8
    2.08
    6
    2.08
    c
    2.06
    7
    1.92
    f
    1.86
    section
    1.84
    me
    1.82
    Activations Density 0.022%

    No Known Activations