© 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. 14432
    Prev
    Next
    INDEX
    Explanations

    image contexts

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

    The neuron strongly activates on the token “image” (e.g. XML/HTML tags, import names, or identifiers containing “image”).

    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
     monstrous
    -2.19
     to
    -2.14
    苤
    -2.14
     captivating
    -2.14
     relentless
    -2.14
     Of
    -2.11
     sooo
    -2.08
    再加上
    -2.06
    っています
    -2.02
     
    -1.98
    POSITIVE LOGITS
    </i>
    2.67
     برای
    2.50
     untuk
    2.45
    Before
    2.36
    when
    2.34
    𖡼
    2.34
    Ⲉ
    2.27
    ⟐
    2.27
    Being
    2.25
    ورك
    2.25
    Activations Density 0.007%

    No Known Activations