© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 66856
    Prev
    Next
    INDEX
    Explanations

    findings from analyses

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

    This neuron fires strongly on single-word scientific section headings or method/action terms (e.g. “Results,” “Analysis,” “Determination,” “Addition”) that typically introduce sections in research papers.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
    其中
    -1.09
    mdash
    -1.08
    aneity
    -1.07
    (!__
    -1.05
    Demografie
    -0.98
    被称为
    -0.97
    //.
    -0.96
    ‌.
    -0.96
    ː
    -0.93
     женский
    -0.92
    POSITIVE LOGITS
     reveals
    1.45
     showed
    1.34
     reveal
    1.29
     suggests
    1.24
     shows
    1.23
     indicate
    1.17
     our
    1.16
     indicates
    1.16
     into
    1.12
     suggest
    1.12
    Activations Density 0.149%

    No Known Activations