© 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. 10-GEMMASCOPE-RES-131K
    7. 11479
    Prev
    Next
    INDEX
    Explanations

    IPCC computer models

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

    The neuron fires on isolated non‐alphanumeric symbol tokens—things like HTML/conditional‐comment delimiters, singleton operators or punctuation (e.g. “<!--[”, “±”, “\lfloor”), and other formatting or math symbols.

    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
    蕯
    -2.39
     твое
    -2.36
    いろいろな
    -2.34
    雰
    -2.31
     Presently
    -2.28
    たとえば
    -2.25
    上官
    -2.20
    驀
    -2.19
    If
    -2.14
    各类
    -2.14
    POSITIVE LOGITS
    }
    3.59
    \
    2.63
    }.
    2.61
     been
    2.39
    憷
    2.23
    $.
    2.20
    。
    2.20
    𖤍
    2.14
    ル
    2.14
    『
    2.13
    Activations Density 0.004%

    No Known Activations