© 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. 34-GEMMASCOPE-RES-131K
    7. 33041
    Prev
    Next
    INDEX
    Explanations

    dagger, Ancient, Monday, Yves

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

    The neuron fires on structural or formatting tokens used in outlines and lesson‐style headings—things like colons, numbered labels, parentheses, and other list/section markers.

    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
    BIÉN
    -0.79
     contempla
    -0.77
     Ekonomi
    -0.75
    況
    -0.73
     Aufl
    -0.73
     felpa
    -0.73
    tą
    -0.73
     teto
    -0.72
     china
    -0.72
    zione
    -0.71
    POSITIVE LOGITS
    limus
    0.84
    rawdę
    0.82
    medaille
    0.81
     rattan
    0.80
     Ia
    0.80
    箍
    0.79
    propylene
    0.78
    NullException
    0.78
     LAR
    0.78
     ornamentation
    0.77
    Activations Density 0.014%

    No Known Activations