© 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. 34-GEMMASCOPE-RES-131K
    7. 31241
    Prev
    Next
    INDEX
    Explanations

    Continue reading

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

    The neuron is tuned to spot URL‐style tokens and path fragments (e.g. domain names, slashes, file‐path segments).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     vino
    -0.80
    ánló
    -0.70
    pstmt
    -0.69
    IVEREF
    -0.69
     カバー
    -0.68
    ates
    -0.66
    art
    -0.66
     製造
    -0.66
     CEP
    -0.65
     tenets
    -0.65
    POSITIVE LOGITS
     hozzá
    0.70
    nault
    0.69
     sağlar
    0.64
    
    0.63
     komunik
    0.63
    mployment
    0.62
     hijau
    0.62
    Adapted
    0.61
     Juana
    0.61
     kasarigan
    0.61
    Activations Density 0.059%

    No Known Activations