© 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. 79947
    Prev
    Next
    INDEX
    Explanations

    non-english words

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

    The neuron is effectively measuring each continuous text block’s length, firing in proportion to how many characters (tokens) appear in that block.

    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
     one
    -1.30
    ,
    -1.16
     really
    -1.10
    :
    -1.08
    One
    -1.08
     star
    -1.07
     ‘
    -1.06
     […]
    -1.02
    !--
    -0.99
     first
    -0.99
    POSITIVE LOGITS
    Comté
    1.38
     considérons
    1.23
     北欧
    1.20
     soulign
    1.19
     thier
    1.16
    Vea
    1.14
     också
    1.13
     admiten
    1.12
    なんですね
    1.12
    zusetzen
    1.11
    Activations Density 0.200%

    No Known Activations