© 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. 72496
    Prev
    Next
    INDEX
    Explanations

    error messages and descriptions

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

    The neuron fires on standalone label‐style words (often capitalized) that introduce a section or UI/action prompt—e.g. “Error:”, “Share this:”, “Subscribe:”, “Rating:” etc.

    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
    ！】
    -1.24
     extré
    -1.20
     chanti
    -1.18
    ächtigen
    -1.16
    ),
    -1.16
    AccessException
    -1.12
    }>
    -1.11
     Projet
    -1.09
     gants
    -1.09
    ]\\
    -1.09
    POSITIVE LOGITS
     to
    1.70
    Their
    1.39
     jedynie
    1.31
    espé
    1.30
     should
    1.29
    Only
    1.27
     selepas
    1.26
    ldots
    1.23
     will
    1.23
     has
    1.23
    Activations Density 0.059%

    No Known Activations