© 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. 67565
    Prev
    Next
    INDEX
    Explanations

    most followed by descriptive words

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

    The neuron fires on emphatic degree words and superlative intensifiers—especially “most” and similarly strong adjectives or adverbs.

    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
     actualiza
    -1.36
    बा
    -1.32
    ização
    -1.32
    ⢅
    -1.32
    ೢ
    -1.31
    忖
    -1.30
    屄
    -1.30
    下さ
    -1.30
    岨
    -1.29
    泂
    -1.29
    POSITIVE LOGITS
    While
    2.06
    They
    1.93
    There
    1.91
    Although
    1.85
    What
    1.72
    It
    1.68
    Another
    1.64
    Very
    1.60
    1
    1.59
    After
    1.56
    Activations Density 0.014%

    No Known Activations