© 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
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 47682
    Prev
    Next
    INDEX
    Explanations

    So I have

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

    This neuron flags first-person self-references (tokens like “I,” “I’m,” “I’ve,” and “my”).

    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
     facilidad
    -1.52
     FIVE
    -1.45
     abundancia
    -1.45
    に入れ
    -1.42
     каждой
    -1.42
     promover
    -1.41
     flera
    -1.39
     déterminé
    -1.39
    噠
    -1.38
    менить
    -1.37
    POSITIVE LOGITS
     Essentially
    1.51
     undeniably
    1.45
    　（
    1.38
    𖣘
    1.34
     about
    1.34
    次回は
    1.31
     ostensibly
    1.30
    Going
    1.30
     eterna
    1.29
     endearing
    1.28
    Activations Density 0.051%

    No Known Activations