© 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. 22-GEMMASCOPE-RES-131K
    7. 67842
    Prev
    Next
    INDEX
    Explanations

    suggest trying this

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

    The neuron fires on phrases offering a solution or advice—e.g. “Try setting … to true in web.xml” or “I think I found something that might help you.”

    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
    %[
    -0.93
    だからこそ
    -0.93
    snip
    -0.92
     MutableLiveData
    -0.89
    mMap
    -0.87
    mallow
    -0.86
    flop
    -0.86
    quations
    -0.84
    Voltar
    -0.82
     指輪
    -0.82
    POSITIVE LOGITS
     your
    1.56
     try
    1.32
     попробу
    1.20
     hopefully
    1.16
     you
    1.13
    试试
    1.12
     thử
    1.10
     will
    1.09
     then
    1.07
    try
    1.05
    Activations Density 0.059%

    No Known Activations