© 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. 22-GEMMASCOPE-RES-131K
    7. 54370
    Prev
    Next
    INDEX
    Explanations

    scientific and academic concepts

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

    the first word of an abstract or academic document.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    The neuron fires strongly on the very first content words at the start of each abstract or section (i.e. the initial key term that opens a new abstract paragraph).

    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
    蛱
    -1.48
     图库
    -1.45
    靦
    -1.36
    龇
    -1.35
    Kde
    -1.34
    upported
    -1.32
     薬局
    -1.30
    ſſen
    -1.30
    嘮
    -1.28
    lorn
    -1.27
    POSITIVE LOGITS
     S
    1.52
     X
    1.46
     has
    1.45
     C
    1.41
     often
    1.38
     L
    1.38
     T
    1.35
    !
    1.34
     This
    1.33
     Since
    1.32
    Activations Density 0.644%

    No Known Activations