© 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. 10-GEMMASCOPE-RES-131K
    4. 57462
    Prev
    Next
    INDEX
    Explanations

    through the roof, way cheaper

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

    The neuron identifies hyperbolic intensifier phrases (e.g. “through the roof,” “boat loads,” “over-the-top,” “mother of all”) that signal extreme degree or magnitude.

    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
     our
    -1.84
     vägen
    -1.33
    ma
    -1.30
    6
    -1.27
    9
    -1.27
     Aufl
    -1.24
    by
    -1.22
     våra
    -1.22
    st
    -1.20
     encontra
    -1.20
    POSITIVE LOGITS
    气氛
    1.61
    nessy
    1.44
    atasaray
    1.38
    手绘
    1.35
    They
    1.35
     svet
    1.33
    发了
    1.32
     prestation
    1.31
     idée
    1.30
    鈁
    1.29
    Activations Density 0.097%

    No Known Activations