© 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. 39614
    Prev
    Next
    INDEX
    Explanations

    ITS and neuronal activity

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

    acronyms and technical terms like "ITS", "Hermione", and words related to neuronal or biological systems.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    The neuron fires on named entities and acronyms—i.e. capitalized proper names (like “Hermione”) and uppercase initialisms (like “ITS”).

    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
    </em>
    -2.23
    \_
    -2.17
     مختلفة
    -2.11
     utilisée
    -2.11
    </h6>
    -2.06
    違って
    -2.06
    蠃
    -2.06
    愎
    -2.00
    chevêque
    -1.98
    ）-
    -1.96
    POSITIVE LOGITS
    al
    3.36
    ra
    2.78
    c
    2.63
    because
    2.63
    re
    2.52
     soo
    2.47
    {
    2.47
    u
    2.47
     hohe
    2.41
     journalist
    2.38
    Activations Density 0.004%

    No Known Activations