© 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. 48596
    Prev
    Next
    INDEX
    Explanations

    code initialization, organizational roles, event titles

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

    This neuron fires on capitalized tokens that are part of names, titles, or other proper‐noun labels.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    technical or programming-related text, particularly code snippets and structured data formats.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @jyhe0408

    capitalized standalone words that look like headers, keywords, or emphasized terms (often at line starts or within code/metadata).

    oai_token-act-pair · gpt-5Triggered 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
     from
    -1.27
     with
    -1.12
     at
    -1.07
     through
    -1.06
     produced
    -1.04
     within
    -0.97
     that
    -0.96
     for
    -0.96
     including
    -0.93
     during
    -0.90
    POSITIVE LOGITS
    vermectin
    1.16
    ebly
    1.08
    existsSync
    1.05
     horrid
    1.05
    熜
    1.02
     comprendre
    1.02
     больших
    1.00
     grec
    0.99
    brtc
    0.99
    -[
    0.96
    Activations Density 0.000%

    No Known Activations