© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 59353
    Prev
    Next
    INDEX
    Explanations

    html view helpers

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

    This neuron primarily fires on code-related tokens—especially names of programming frameworks, libraries, and other technical identifiers.

    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
     Melvin
    -0.81
     Viscount
    -0.77
    銃
    -0.76
     Schme
    -0.73
    izzard
    -0.72
    TypeEnum
    -0.71
    ɻ
    -0.71
    ⇎
    -0.70
    StatusEnum
    -0.69
    履き
    -0.69
    POSITIVE LOGITS
     helper
    2.19
     helpers
    2.17
    helpers
    2.02
    Helper
    1.93
     Helper
    1.82
    helper
    1.80
     Helpers
    1.63
    Helpers
    1.59
     helping
    1.42
     view
    1.29
    Activations Density 0.113%

    No Known Activations