© 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. 85321
    Prev
    Next
    INDEX
    Explanations

    citations like [@B1]

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

    This neuron activates on tokens composed of (or consisting entirely of) uppercase letters or uppercase‐letter sequences.

    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
     geniales
    -1.70
     motivadoras
    -1.66
     pudo
    -1.65
     primeiras
    -1.65
     how
    -1.62
     and
    -1.59
     segura
    -1.55
     schnell
    -1.52
    </h5>
    -1.52
     '+':
    -1.48
    POSITIVE LOGITS
    .
    1.76
    like
    1.70
     apparaissent
    1.63
    歺
    1.63
     jiné
    1.63
    who
    1.59
    pped
    1.58
     marie
    1.57
     roland
    1.55
    睃
    1.51
    Activations Density 0.059%

    No Known Activations