© 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
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 14251
    Prev
    Next
    INDEX
    Explanations

    sports record or statistics

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

    The neuron fires on numeric sequences with dashes or slashes—e.g. win-loss records, scores or stat lines.

    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
    おわりに
    -1.22
    の準備
    -1.19
    ことが
    -1.19
    b
    -1.16
    つづく
    -1.16
    ほら
    -1.14
    様が
    -1.09
     какой
    -1.06
    えっ
    -1.05
    明日は
    -1.05
    POSITIVE LOGITS
    <bos>
    1.35
    ariConfig
    1.22
    があります
    1.21
    astify
    1.16
    かれ
    1.15
    省市镇
    1.13
     yatak
    1.12
    そして
    1.11
     作り方
    1.10
     including
    1.09
    Activations Density 0.001%

    No Known Activations