© 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. 10-GEMMASCOPE-RES-131K
    7. 41077
    Prev
    Next
    INDEX
    Explanations

    Jerry Brown

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

    The neuron fires strongly on substrings of multi-syllable capitalized names or titles (i.e. parts of proper nouns).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
    Ꮓ
    -1.63
    \
    -1.51
    schild
    -1.42
    deepcopy
    -1.41
    Ⓡ
    -1.39
    Ᏸ
    -1.39
    ії
    -1.38
    ization
    -1.37
    ጴ
    -1.36
    <0xE3>
    -1.33
    POSITIVE LOGITS
    訁
    1.66
    ウォーター
    1.58
     algunas
    1.58
     again
    1.52
     still
    1.52
    锵
    1.50
     my
    1.48
     me
    1.47
     ahora
    1.46
    かなりの
    1.45
    Activations Density 0.053%

    No Known Activations