© 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. 57178
    Prev
    Next
    INDEX
    Explanations

    numbers ending in en or th

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

    The neuron activates strongly on specialized technical or domain-specific terms—e.g. medical procedure words (biliary, stent, hepat) or software/markup jargon (JavaScript, ProgId, draft)—so it’s looking for highly technical vocabulary.

    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
     eighty
    -1.40
     Eighty
    -1.37
     sixty
    -1.25
     seventy
    -1.24
     Sixty
    -1.17
     Seventy
    -1.16
     Ninety
    -1.13
     ninety
    -1.04
     fifty
    -1.01
     Fifty
    -0.96
    POSITIVE LOGITS
    Thirteen
    1.02
     Thirteen
    0.93
     Six
    0.84
     SEVEN
    0.82
    seventh
    0.82
    Six
    0.81
     six
    0.80
     IX
    0.78
     sieben
    0.77
    Seventh
    0.77
    Activations Density 0.228%

    No Known Activations