© 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
    2. Gemma-2-27B
    3. 34-GEMMASCOPE-RES-131K
    4. 83237
    Prev
    Next
    INDEX
    Explanations

    going off on tangents

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

    The neuron strongly fires on discourse-marking words and punctuation (e.g. “But,” “However,” “digress,” colons, “Without further ado,” etc.) that signal topic shifts or commentary asides.

    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
     menyus
    -0.90
     Buren
    -0.79
     Schriftsteller
    -0.76
    rieta
    -0.73
    ียม
    -0.72
    tadores
    -0.71
    cien
    -0.71
     namorados
    -0.70
    ITARY
    -0.70
    homepage
    -0.69
    POSITIVE LOGITS
     off
    3.00
     digress
    2.97
     tangent
    2.34
     tangents
    2.22
    Off
    2.13
     detour
    2.05
     tangential
    2.03
    off
    2.00
     diversion
    2.00
     Off
    1.96
    Activations Density 0.053%

    No Known Activations