© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 82651
    Prev
    Next
    INDEX
    Explanations

    person names, biological origins, colors

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

    color values or color-related parameters in technical or programming contexts.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    The neuron activates on color‐specification tokens in code or markup (e.g. numeric color values and named colors).

    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
    namespaces
    -0.76
     Độ
    -0.76
     helg
    -0.74
    fluid
    -0.73
     walked
    -0.72
    riendo
    -0.72
     with
    -0.71
     content
    -0.71
     fluid
    -0.71
    rough
    -0.70
    POSITIVE LOGITS
     vermelho
    0.85
    tering
    0.79
    yaa
    0.77
    inspiration
    0.75
     empress
    0.74
    Споља
    0.73
     red
    0.73
     noir
    0.73
    chang
    0.72
    ⡠
    0.72
    Activations Density 0.000%

    No Known Activations