© 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
    2. Gemma-2-27B
    3. 34-GEMMASCOPE-RES-131K
    4. 108532
    Prev
    Next
    INDEX
    Explanations

    lewin, norton, marilyn, veloster, nsg, plumber, sarcasm

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

    The neuron responds to subword fragments of URLs and email/domain names.

    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
    などは
    -1.26
    などの
    -1.23
    センス
    -1.12
     in
    -1.10
    を
    -1.09
    などを
    -1.09
    があり
    -1.04
    をはじめ
    -1.00
    などが
    -1.00
    があります
    -1.00
    POSITIVE LOGITS
    為に
    1.86
    事も
    1.41
    ところに
    1.38
    cobacterium
    1.38
    んですよ
    1.37
    ところは
    1.31
    感じで
    1.30
    isait
    1.30
     giud
    1.29
    んですよね
    1.29
    Activations Density 0.085%

    No Known Activations