© 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. 28413
    Prev
    Next
    INDEX
    Explanations

    date/time stamps

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

    The neuron strongly activates on capitalized heading or title words—i.e. tokens that occur in section titles, metadata lines, or other title‐cased headings.

    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
     September
    -1.20
     April
    -1.16
     July
    -1.09
     October
    -1.06
     January
    -1.05
     December
    -1.05
     February
    -1.04
     March
    -1.01
    G
    -1.00
     June
    -0.99
    POSITIVE LOGITS
     kalori
    1.11
     then
    1.06
    んじゃ
    1.03
     tacón
    1.03
    γή
    1.03
    galkan
    1.00
     opět
    0.99
    𝕸
    0.99
    Kommentare
    0.99
    giày
    0.97
    Activations Density 0.215%

    No Known Activations