© 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. 22-GEMMASCOPE-RES-131K
    4. 12904
    Prev
    Next
    INDEX
    Explanations

    ingestion and swallowing

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

    words related to consumption or intake, particularly "ingestion" and "swallowing".

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

    The main thing this neuron does is spot references to ingesting or consuming substances (e.g., words like “ingestion,” “swallowing,” “drinking,” “eat”).

    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
     you
    -1.50
    and
    -1.36
     соответствует
    -1.33
    on
    -1.31
    in
    -1.27
    \
    -1.27
     \
    -1.25
     которую
    -1.25
     другие
    -1.25
    f
    -1.23
    POSITIVE LOGITS
     Öffnen
    1.61
     intake
    1.57
    ܜ
    1.48
     consume
    1.38
    ulur
    1.36
     største
    1.30
    пасибо
    1.27
     ingest
    1.27
     borracha
    1.27
    acijos
    1.27
    Activations Density 0.032%

    No Known Activations