© 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
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 84876
    Prev
    Next
    INDEX
    Explanations

    HTTP response writing methods

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

    The neuron strongly responds to programming‐language punctuation and syntax tokens (e.g. dots, parentheses, braces, semicolons) in C# code snippets.

    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
    Alain
    -0.69
     verständ
    -0.68
     Хоро
    -0.68
     besk
    -0.66
     telescopic
    -0.66
    selectedIndex
    -0.64
    𝘀
    -0.64
     Wtf
    -0.63
     мире
    -0.62
     gặp
    -0.62
    POSITIVE LOGITS
     Response
    2.50
    Response
    2.48
     response
    2.05
     RESPONSE
    1.76
    response
    1.74
    RESPONSE
    1.59
     réponse
    1.46
     Server
    1.44
     Responses
    1.38
    Server
    1.34
    Activations Density 0.098%

    No Known Activations