© 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. 34-GEMMASCOPE-RES-131K
    7. 57397
    Prev
    Next
    INDEX
    Explanations

    best western

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

    The neuron consistently fires on numeric tokens (particularly multi‐digit numbers, such as years, IDs, or other numerical values).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     peti
    -0.85
     epistem
    -0.82
    Fantasy
    -0.81
     glen
    -0.81
    偽
    -0.78
     Atlan
    -0.73
     EO
    -0.73
    $_
    -0.73
     diec
    -0.73
    Epide
    -0.73
    POSITIVE LOGITS
     colher
    0.85
     riservata
    0.84
    📘
    0.81
     Better
    0.79
    ösen
    0.78
    это
    0.77
    ERRA
    0.77
    Produits
    0.76
    jobb
    0.76
    是什么
    0.76
    Activations Density 0.023%

    No Known Activations