© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 84262
    Prev
    Next
    INDEX
    Explanations

    self-

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

    The neuron fires on the fragmented prefix “self-,” i.e. tokens that are part of words beginning with “self-.”

    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
     it
    -2.77
     The
    -2.55
     But
    -2.44
    );
    -2.30
    .
    -2.28
     Then
    -2.22
     Our
    -2.09
     -
    -2.05
     Who
    -1.99
     its
    -1.90
    POSITIVE LOGITS
    ୌ
    2.53
    ︕
    2.50
    ୈ
    2.45
    𓆜
    2.44
     estadounid
    2.44
    ⢓
    2.41
    ﮏ
    2.34
    
    2.33
    橚
    2.33
    綃
    2.31
    Activations Density 0.012%

    No Known Activations