© 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. 54158
    Prev
    Next
    INDEX
    Explanations

    tax law

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

    The neuron strongly responds to tokens that mark the start of new sections or list items—i.e. headings, instruction labels, and other structural markers in the text.

    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
    bases
    -0.81
    ][$
    -0.77
    burst
    -0.75
     ".
    -0.73
     essas
    -0.72
    (/^
    -0.71
     municipale
    -0.70
    }$$
    -0.69
     dłu
    -0.69
     Ju
    -0.69
    POSITIVE LOGITS
     setter
    1.09
     setters
    1.02
    汁
    0.93
    Indicator
    0.88
    Cry
    0.86
    indica
    0.84
     cry
    0.84
     parsing
    0.83
    annotations
    0.83
    cry
    0.81
    Activations Density 0.005%

    No Known Activations