© 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. 1198
    Prev
    Next
    INDEX
    Explanations

    mean

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

    This neuron fires on programming‐style identifiers and technical tokens (e.g. code keywords, file‐extension names, variable names with underscores) rather than natural language.

    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
    豚肉
    -0.80
    minded
    -0.79
    penta
    -0.76
    coring
    -0.74
    pages
    -0.74
    (?
    -0.74
    heartedly
    -0.73
    iteter
    -0.72
     penta
    -0.72
    aris
    -0.72
    POSITIVE LOGITS
     mean
    1.69
    mean
    1.33
    MEAN
    1.30
    Mean
    1.29
     Mean
    1.25
     MEAN
    1.01
    Means
    0.99
    eig
    0.97
    جاد
    0.91
     means
    0.90
    Activations Density 0.003%

    No Known Activations