© 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
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 59061
    Prev
    Next
    INDEX
    Explanations

    CO2

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

    The chemical formula CO2 (carbon dioxide) and related terminology.

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

    The neuron flags occurrences of the uppercase “CO” token (as in chemical formulas like CO₂), i.e. mentions of the carbon‐oxygen compound notation.

    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
    which
    -3.25
    up
    -2.88
     Importantly
    -2.80
    where
    -2.72
    most
    -2.55
    of
    -2.47
    ri
    -2.44
    please
    -2.42
    c
    -2.39
    炻
    -2.39
    POSITIVE LOGITS
     has
    2.77
    т
    2.27
    ارتمان
    2.25
     {
    2.22
    犼
    2.20
    ৣ
    2.20
     другим
    2.19
     adventurous
    2.17
    廨
    2.17
    秝
    2.16
    Activations Density 0.009%

    No Known Activations