© 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. 68155
    Prev
    Next
    INDEX
    Explanations

    IR spectroscopy and abbreviations

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

    abbreviations IR (infrared/information retrieval) in technical contexts.

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

    The neuron consistently fires on the token “IR,” i.e. occurrences of the abbreviation IR (often in contexts like IR spectroscopy or IR followed by numbers).

    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
    寕
    -2.63
     небольшой
    -2.56
     resembling
    -2.53
    菹
    -2.52
     verschied
    -2.44
    _{
    -2.41
     مي‌
    -2.39
     relativi
    -2.39
    <bos>
    -2.36
    with
    -2.36
    POSITIVE LOGITS
    ization
    2.64
     has
    2.44
     sportowe
    2.31
     ważne
    2.20
    }
    2.17
    kaç
    2.16
    🖐
    2.14
    ❣
    2.11
    [
    2.11
    çok
    2.08
    Activations Density 0.007%

    No Known Activations