© 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. 22-GEMMASCOPE-RES-131K
    4. 69160
    Prev
    Next
    INDEX
    Explanations

    identifying questions

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

    The neuron flags interview question prompts or headings (the “Q:”/“What …?” lines).

    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
     also
    -1.95
     those
    -1.19
     with
    -1.09
     will
    -1.09
    그리고
    -1.08
     також
    -1.04
    мія
    -1.03
     other
    -1.02
     berikutnya
    -1.02
     future
    -1.00
    POSITIVE LOGITS
     how
    1.18
    そも
    1.12
     starting
    1.11
     mantenerse
    1.06
     cuarenta
    1.05
     как
    1.04
     Describe
    1.03
    普通に
    1.02
    nitts
    1.02
     How
    1.02
    Activations Density 0.195%

    No Known Activations