© 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. 22377
    Prev
    Next
    INDEX
    Explanations

    voice

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

    The neuron consistently fires on occurrences of the specific token “Voice.”

    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
    -3.39
     Which
    -3.20
     
    -2.80
    (
    -2.78
    ...
    -2.59
    et
    -2.42
     einzigen
    -2.41
     I
    -2.39
     â
    -2.39
    嵛
    -2.34
    POSITIVE LOGITS
    獁
    3.30
    และ
    2.81
    蕷
    2.67
    ꯶
    2.67
    灕
    2.66
    ❗️
    2.63
    Ruju
    2.61
     剤
    2.59
    providedIn
    2.58
     动漫
    2.48
    Activations Density 0.007%

    No Known Activations