© 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
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 66141
    Prev
    Next
    INDEX
    Explanations

    prefixes of compound words

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

    The neuron fires on scientific-style stems and prefixes—especially technical jargon or compound-word roots (e.g. poly-, semi-, non-, response, concentration, carbon, drop, flex, left)—marking formal academic terminology.

    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
     with
    -1.47
     both
    -1.45
     but
    -1.44
     other
    -1.34
     any
    -1.32
     others
    -1.27
     all
    -1.24
     many
    -1.16
     mostly
    -1.13
     on
    -1.13
    POSITIVE LOGITS
     logiciels
    1.31
    Только
    1.21
    पि
    1.17
     indescri
    1.16
     allan
    1.14
     pérd
    1.13
    Tél
    1.13
    ificato
    1.13
     muszą
    1.10
     предлагает
    1.09
    Activations Density 0.120%

    No Known Activations