© 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. 22-GEMMASCOPE-RES-131K
    7. 104449
    Prev
    Next
    INDEX
    Explanations

    purpose of action

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

    The neuron fires on mentions of using or connecting to specific software libraries or packages.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
     something
    -1.21
     verdaderas
    -1.14
     certain
    -1.09
     either
    -1.08
     ciertos
    -1.06
     Either
    -1.05
    🥽
    -1.05
     tons
    -1.04
     ciertas
    -1.04
     mismas
    -1.03
    POSITIVE LOGITS
     chocs
    1.11
     truff
    1.10
     both
    1.10
     jong
    1.10
     både
    1.09
     castro
    1.08
     kilom
    1.06
     marron
    1.06
     COSTA
    1.06
    今回は
    1.04
    Activations Density 0.108%

    No Known Activations