© 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. 64100
    Prev
    Next
    INDEX
    Explanations

    names or specific entities

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

    The neuron is picking up on programming‐style formatting (code snippets)—it fires strongly on things like braces, indentation/tabs, camel-case identifiers, and other code punctuation rather than ordinary prose.

    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
     Lorenzo
    -0.71
     pleaded
    -0.70
     welches
    -0.67
     originate
    -0.66
     mayoría
    -0.66
     bis
    -0.66
    llis
    -0.66
     Laus
    -0.65
     hyper
    -0.65
     boreal
    -0.65
    POSITIVE LOGITS
    Crawler
    0.80
    skilled
    0.78
    DRC
    0.75
    staking
    0.75
    的选择
    0.75
    ūd
    0.74
    tokine
    0.73
     retratos
    0.73
     szí
    0.73
    ulative
    0.73
    Activations Density 0.001%

    No Known Activations