© 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. 130985
    Prev
    Next
    INDEX
    Explanations

    The neuron strongly activates on personal or character names—that is, capitalized named entities referring to people or fictional characters.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    names and titles

    np_acts-logits-general · gemini-2.5-flash-lite
    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
     stod
    -1.10
    扣
    -1.06
    に似
    -1.06
     italienne
    -1.02
    に入れる
    -1.02
    バナナ
    -0.98
    ところに
    -0.96
     goddamn
    -0.95
    ちょい
    -0.94
     этом
    -0.94
    POSITIVE LOGITS
    ༵
    1.16
    𓍊
    1.08
    <bos>
    1.06
    =}
    1.05
    aliases
    1.04
    akad
    1.04
     nové
    1.03
     himself
    1.03
     who
    1.02
    achter
    0.98
    Activations Density 0.065%

    No Known Activations