© 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. 84772
    Prev
    Next
    INDEX
    Explanations

    probes another, tout stimulus

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

    The neuron fires on verbs that express a speaker’s stance or evaluation (e.g., opinion or sentiment verbs like “admire,” “disparage,” “decried,” “mock,” etc.).

    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
     in
    -2.70
     a
    -2.59
    </h3>
    -2.28
     now
    -2.19
     "
    -2.13
     When
    -2.09
     into
    -2.08
     D
    -2.03
     more
    -2.03
    </h2>
    -1.99
    POSITIVE LOGITS
    錞
    2.36
     手绘
    2.31
     pérd
    2.28
    ௅
    2.23
     podoba
    2.20
    秶
    2.20
     كيفية
    2.14
     gardien
    2.13
    鳚
    2.11
     kablo
    2.09
    Activations Density 0.189%

    No Known Activations