© 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. 22-GEMMASCOPE-RES-131K
    7. 98310
    Prev
    Next
    INDEX
    Explanations

    codes, citations, and identifiers

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

    The neuron activates on four‐digit numeric tokens—especially years or year‐like identifiers.

    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
     blusa
    -1.63
     pelu
    -1.48
     blusas
    -1.44
     régi
    -1.43
     média
    -1.43
    鼕
    -1.42
     laiton
    -1.42
     eenen
    -1.41
     botas
    -1.41
     män
    -1.41
    POSITIVE LOGITS
    f
    1.79
    n
    1.69
    A
    1.67
    w
    1.63
    g
    1.61
    Y
    1.61
    L
    1.59
    C
    1.58
     also
    1.55
    b
    1.51
    Activations Density 0.001%

    No Known Activations