© 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. 67469
    Prev
    Next
    INDEX
    Explanations

    Abra- and abrasive- words

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

    This neuron detects occurrences of the character sequence “abra” (in any word containing that substring).

    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
    
    -3.05
    na
    -2.48
    .”
    -2.47
    潮流
    -2.23
    нні
    -2.22
    ve
    -2.22
    <td>
    -2.22
     ‘’
    -2.20
    at
    -2.19
     whats
    -2.19
    POSITIVE LOGITS
    咴
    3.61
    ཋ
    2.80
    沨
    2.78
    
    2.78
    蟶
    2.75
    咵
    2.73
    鱨
    2.72
    葙
    2.70
    耵
    2.66
    
    2.64
    Activations Density 0.003%

    No Known Activations