© 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. 10-GEMMASCOPE-RES-131K
    7. 73900
    Prev
    Next
    INDEX
    Explanations

    Cath and Cong prefixes

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

    The neuron consistently fires on the letter sequence “Cath,” i.e. tokens beginning with “Cath.”

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
     monumental
    -2.47
    久し
    -2.39
     The
    -2.34
     englischen
    -2.33
     traditionellen
    -2.30
     achten
    -2.28
    勿論
    -2.28
    媗
    -2.27
     dritten
    -2.27
     Так
    -2.27
    POSITIVE LOGITS
    "
    3.16
    ’
    2.97
    r
    2.80
    it
    2.73
     δεν
    2.61
    and
    2.52
    om
    2.48
    𝐉
    2.44
     didnt
    2.42
    are
    2.39
    Activations Density 0.008%

    No Known Activations