© 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. 34-GEMMASCOPE-RES-131K
    7. 60796
    Prev
    Next
    INDEX
    Explanations

    documentation and code examples

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

    The neuron fires on technical download/install jargon—especially file‐format names (like “iso” or “.dmg”), version‐number tokens, OS names (e.g. “Windows”), and similar installation‐related terms.

    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
     היתה
    -1.27
     delivers
    -1.16
     when
    -1.16
     eight
    -1.15
     delivering
    -1.12
    𝒙
    -1.11
     in
    -1.11
     at
    -1.10
    hdashline
    -1.10
     then
    -1.06
    POSITIVE LOGITS
     ≈
    1.31
    ⨾
    1.28
    Также
    1.27
    优秀的
    1.25
    🩹
    1.23
    Alright
    1.20
    Woah
    1.17
    Advertisements
    1.16
     fuf
    1.15
    кож
    1.15
    Activations Density 0.217%

    No Known Activations