© 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-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 2858
    Prev
    Next
    INDEX
    Explanations

    built for specific purpose

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

    The neuron activates on standalone title-style headings or list-item labels (short, title-cased phrases announcing a section or point).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    capitalized words that begin titles, headings, or sections in informational or instructional text.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @jyhe0408

    capitalized function words within title-case headings or section titles.

    oai_token-act-pair · gpt-5Triggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-12b-pt/resid_post/layer_24_width_16k_l0_medium
    Prompts (Dashboard)
    392,802 prompts, 256 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     rigidbody
    1.07
     restable
    1.01
    carpeta
    0.98
    pyrazol
    0.96
    classpath
    0.95
     deserial
    0.95
    snackbar
    0.95
     conformations
    0.94
    enseignement
    0.94
    piperidin
    0.94
    POSITIVE LOGITS
     L
    1.45
     M
    1.40
     C
    1.38
     U
    1.34
     K
    1.30
     B
    1.29
     D
    1.29
     S
    1.28
     G
    1.28
    D
    1.28
    Activations Density 0.000%

    No Known Activations