© 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. 16128
    Prev
    Next
    INDEX
    Explanations

    technical documentation headers

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

    The neuron activates on Markdown second‐level headings that consist solely of “##” followed by a numeric identifier.

    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
    1
    -3.19
    0
    -2.81
     formidable
    -2.41
     visitas
    -2.27
     enormous
    -2.22
     fairytale
    -2.20
    6
    -2.19
    ↵↵
    -2.14
     Bedürfn
    -2.05
    もありました
    -2.05
    POSITIVE LOGITS
    ti
    2.58
    ра
    2.55
    ⢊
    2.31
    j
    2.31
    ☑
    2.27
    le
    2.27
     armen
    2.25
    k
    2.25
     priva
    2.22
    ⁉
    2.22
    Activations Density 0.007%

    No Known Activations