© 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. 34-GEMMASCOPE-RES-131K
    7. 117294
    Prev
    Next
    INDEX
    Explanations

    constraint satisfaction problem solving

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

    The neuron strongly fires on code- or research-style technical jargon—especially programming and licensing terms (e.g. “constraint,” “logic,” “Version 2.0,” “forward,” “checking,” 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_34/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
     getTitle
    -0.73
    }_{-
    -0.71
     CODES
    -0.68
     fogo
    -0.67
     Merkel
    -0.66
     clientèle
    -0.66
    superuser
    -0.65
     Cea
    -0.65
     Cupcakes
    -0.65
     دسته
    -0.64
    POSITIVE LOGITS
    Domain
    1.13
     Domain
    1.04
     domain
    1.03
     propag
    1.00
     domains
    0.98
     CSP
    0.95
    CSP
    0.92
     propagation
    0.92
    domain
    0.91
     filtering
    0.91
    Activations Density 0.008%

    No Known Activations