© 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. 10-GEMMASCOPE-RES-131K
    7. 65759
    Prev
    Next
    INDEX
    Explanations

    narrow openings or cuts

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

    The neuron fires on terms naming mechanical openings or cut‐outs (slots, grooves, slits, cuts) in a technical description.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
     abordar
    -1.59
     atraer
    -1.57
     apoyar
    -1.44
    罴
    -1.43
     expandir
    -1.42
     convencer
    -1.40
     optimizar
    -1.36
     rádio
    -1.34
     discutir
    -1.33
     fortalecer
    -1.30
    POSITIVE LOGITS
    '
    1.44
     between
    1.41
     as
    1.40
    CHREIB
    1.36
     so
    1.26
    inerario
    1.16
    市场的
    1.16
     [...]
    1.16
     then
    1.13
     in
    1.12
    Activations Density 0.080%

    No Known Activations