© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 91930
    Prev
    Next
    INDEX
    Explanations

    contcontribute

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

    The neuron detects occurrences of the character sequence “cont” (e.g. in words like contour, contours, contato, contrib, 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_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
    vezet
    -0.84
     columns
    -0.81
    parole
    -0.81
     occupants
    -0.81
     publicKey
    -0.79
    wendig
    -0.79
    
    -0.77
    uvw
    -0.76
    Komple
    -0.76
    BuildConfig
    -0.75
    POSITIVE LOGITS
    cont
    1.68
     cont
    1.63
    CONT
    1.61
     Cont
    1.58
     CONT
    1.56
    Cont
    1.48
     kont
    1.33
     Kont
    1.16
     conte
    1.12
    contour
    1.07
    Activations Density 0.015%

    No Known Activations