© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 40673
    Prev
    Next
    INDEX
    Explanations

    Portable info

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

    This neuron detects title‐style headings or section headers (i.e. capitalized words and line-start/end markers that form titles).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
     dezelve
    -2.00
     eene
    -1.68
     sèche
    -1.63
    釹
    -1.60
     aluminio
    -1.59
     hospod
    -1.58
     eenen
    -1.57
     hunne
    -1.56
    
    -1.55
    眨眼
    -1.55
    POSITIVE LOGITS
    ↵
    1.81
     donde
    1.76
     ).
    1.72
    ’，
    1.65
    不断
    1.59
     algunos
    1.58
     pikir
    1.56
     *
    1.53
     continuó
    1.53
     [
    1.52
    Activations Density 0.008%

    No Known Activations