© 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
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 98570
    Prev
    Next
    INDEX
    Explanations

    programming and system paths

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

    The neuron fires on distinctive multi-character identifier tokens—things like module or library names, function names, variable names, or domain names in code rather than everyday words.

    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
     other
    -1.52
     even
    -1.40
     without
    -1.34
     any
    -1.27
     before
    -1.25
    ॄ
    -1.24
     all
    -1.22
     wakil
    -1.20
     always
    -1.19
     deviceId
    -1.16
    POSITIVE LOGITS
     vaše
    1.43
     naše
    1.33
    ブラシ
    1.32
    ल्ड
    1.30
     жк
    1.28
    hörer
    1.27
    まれた
    1.27
    mmation
    1.27
    PushMatrix
    1.26
    युक्त
    1.22
    Activations Density 0.001%

    No Known Activations