© 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. 60500
    Prev
    Next
    INDEX
    Explanations

    dividing numbers or phrases

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

    The neuron primarily responds to the word “divide” when it appears in mathematical contexts.

    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
    1
    -3.03
    M
    -2.75
    8
    -2.67
    U
    -2.56
     могут
    -2.50
     –
    -2.22
     womit
    -2.19
    ↵↵
    -2.16
    W
    -2.16
    9
    -2.14
    POSITIVE LOGITS
    ݮ
    2.89
    轵
    2.53
    鸸
    2.50
     ſou
    2.50
    tlere
    2.47
    ו
    2.44
    荭
    2.44
    驺
    2.36
    蘩
    2.33
     suéter
    2.30
    Activations Density 0.022%

    No Known Activations