© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 21160
    Prev
    Next
    INDEX
    Explanations

    words indicating source or origin

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

    The neuron fires on occurrences of the preposition “from” (i.e. the token “from” marking a source or origin).

    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
     "
    -3.30
     одной
    -2.88
     "'
    -2.78
     подробно
    -2.75
     этот
    -2.63
     
    -2.55
     rant
    -2.50
    ]."
    -2.48
    <bos>
    -2.45
    ?"
    -2.45
    POSITIVE LOGITS
     afar
    3.14
     They
    3.08
    t
    2.92
     whence
    2.66
     to
    2.61
     from
    2.58
     whom
    2.56
    뻗
    2.56
    इटम
    2.50
     which
    2.44
    Activations Density 0.309%

    No Known Activations