© 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
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 104679
    Prev
    Next
    INDEX
    Explanations

    camera lens specs, musical numbers, years

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

    The neuron fires on tokens that are part of numeric or catalogue‐style specifications—e.g. digits and their adjoining abbreviations (like “50 mm,” “f/1.8,” “Op. 154,” “II”)

    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
    la
    -1.54
     nếu
    -1.48
     he
    -1.48
    D
    -1.46
     菇
    -1.46
    她
    -1.45
    ,”
    -1.45
    are
    -1.44
    );
    -1.44
    這
    -1.41
    POSITIVE LOGITS
    ↵↵
    1.83
    Pgina
    1.44
     "
    1.31
     Crear
    1.30
    usd
    1.30
    logía
    1.29
     lelaki
    1.27
    ѣ
    1.26
     dokter
    1.24
    uccino
    1.24
    Activations Density 0.008%

    No Known Activations