© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 7719
    Prev
    Next
    INDEX
    Explanations

    verified purchase

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

    The neuron activates primarily on document‐level metadata and formatting markers (e.g. ratings, section‐heading underlines of “=”, HTML/XML‐style tag lines).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
     in
    -1.89
     Also
    -1.74
    ！"
    -1.73
     However
    -1.70
     had
    -1.66
     There
    -1.65
    
    -1.60
     gave
    -1.59
     after
    -1.54
     Through
    -1.53
    POSITIVE LOGITS
    镲
    2.13
    ,”
    1.94
    臜
    1.89
    ⚣
    1.80
     纹身
    1.78
     ($("#
    1.78
    淒
    1.73
    allows
    1.66
     kostet
    1.66
     trám
    1.65
    Activations Density 0.000%

    No Known Activations