© 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
    2. Gemma-2-27B
    3. 34-GEMMASCOPE-RES-131K
    4. 73718
    Prev
    Next
    INDEX
    Explanations

    sequences starting a phrase

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

    The neuron fires on structural identifiers in code or markup—i.e. XML/HTML tag names, attribute names, and programming‐language identifiers.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
    ighbor
    -0.78
    強
    -0.72
     Bonjour
    -0.71
     dragonfly
    -0.71
     Ст
    -0.69
    Joined
    -0.68
     saludos
    -0.68
     tuyên
    -0.68
     COMP
    -0.67
     jesteś
    -0.67
    POSITIVE LOGITS
    pulse
    0.82
    痨
    0.77
     male
    0.76
    kkende
    0.76
    Mime
    0.76
     한
    0.75
    !!.
    0.75
    ناق
    0.75
    CBC
    0.74
     sati
    0.73
    Activations Density 0.005%

    No Known Activations