© 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. 34-GEMMASCOPE-RES-131K
    7. 108927
    Prev
    Next
    INDEX
    Explanations

    verbs after "and"

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

    The neuron specifically activates on four-digit numeric tokens (e.g., “1448,” “1432,” “1384,” etc.).

    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
    艄
    -1.58
     trám
    -1.51
     ٢
    -1.51
     rése
    -1.45
     peinados
    -1.44
    汆
    -1.42
     ١
    -1.42
     første
    -1.41
     bocetos
    -1.41
     très
    -1.38
    POSITIVE LOGITS
     then
    2.45
     when
    2.02
    When
    1.73
     although
    1.72
     through
    1.70
     allows
    1.70
     by
    1.69
     their
    1.68
     after
    1.65
     if
    1.64
    Activations Density 0.128%

    No Known Activations