© 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. 22-GEMMASCOPE-RES-131K
    7. 39254
    Prev
    Next
    INDEX
    Explanations

    use NPM to manage

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

    The neuron is selectively active on the tokens that make up the question title or header line (including its key content words and delimiting punctuation).

    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
    常见的
    -1.05
    箭头
    -1.04
     zahlreiche
    -1.01
     hiel
    -1.00
    据说
    -1.00
    前言
    -1.00
     формате
    -0.99
    头部
    -0.97
     fiel
    -0.96
    assemblée
    -0.96
    POSITIVE LOGITS
    はもう
    1.11
     now
    1.09
     enabled
    1.08
    المشار
    1.07
     AIRCRAFT
    1.05
     HEARING
    1.05
     pacient
    1.02
    Archite
    1.01
     confirmó
    1.01
    反而
    1.00
    Activations Density 0.153%

    No Known Activations