© 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. 75189
    Prev
    Next
    INDEX
    Explanations

    what I say

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

    The neuron responds to informal conversational “filler” words and discourse markers (e.g. um, well, I’m not sure) and emphatic punctuation like dashes.

    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
     DISABLE
    -1.19
     fanta
    -1.06
     dispo
    -1.05
     apartament
    -1.03
     Voiture
    -1.02
     nark
    -1.02
    !«
    -1.02
    没有人
    -1.01
     veterinario
    -1.00
     navideños
    -1.00
    POSITIVE LOGITS
     a
    1.33
     an
    1.28
     D
    1.16
     poner
    1.14
     what
    1.13
     sagt
    1.08
     put
    1.07
     called
    1.07
     like
    1.05
     ....
    1.05
    Activations Density 0.015%

    No Known Activations