© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 60756
    Prev
    Next
    INDEX
    Explanations

    ord words and names

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

    The neuron fires on occurrences of the three‐letter sequence “ord,” regardless of where it appears in the text.

    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
    兮
    -2.53
    vantaged
    -2.52
    <bos>
    -2.42
    でしたが
    -2.34
     parachoque
    -2.33
    嗳
    -2.33
    ětí
    -2.28
     médias
    -2.28
     treacherous
    -2.25
     deswegen
    -2.23
    POSITIVE LOGITS
    t
    3.70
    l
    3.53
    j
    2.92
    —
    2.88
    ord
    2.83
    le
    2.83
    '
    2.63
    对
    2.58
    te
    2.48
    2
    2.41
    Activations Density 0.004%

    No Known Activations