© 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. 22-GEMMASCOPE-RES-131K
    7. 25973
    Prev
    Next
    INDEX
    Explanations

    book titles and essays

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

    This neuron activates on bibliographic title text—specifically on book, essay, article, or section titles (often capitalized or in quotes) in reference lists.

    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
     confirmó
    -1.63
     who
    -1.60
     indicó
    -1.57
    ૉ
    -1.52
     informó
    -1.42
     grosir
    -1.38
    omenta
    -1.37
     consente
    -1.34
     vetores
    -1.33
     сообщил
    -1.32
    POSITIVE LOGITS
    PasswordEncoder
    1.40
    mockMvc
    1.39
    şim
    1.35
    ₫
    1.33
     about
    1.30
     on
    1.30
     веке
    1.27
     or
    1.25
    紘
    1.24
     Erziehung
    1.23
    Activations Density 0.028%

    No Known Activations