© 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. 22-GEMMASCOPE-RES-131K
    7. 33164
    Prev
    Next
    INDEX
    Explanations

    associated with object

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

    The neuron preferentially activates on long, multi-syllabic technical or domain-specific 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_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
    などを
    -1.34
     astounding
    -1.30
     Dinas
    -1.23
     strikingly
    -1.22
     astonishing
    -1.17
     alcuni
    -1.16
     häufiger
    -1.13
     wacky
    -1.13
     famed
    -1.07
     méthodes
    -1.07
    POSITIVE LOGITS
     from
    1.68
     what
    1.67
     with
    1.46
     people
    1.36
     without
    1.36
     facilitates
    1.34
     would
    1.33
     nestled
    1.30
     after
    1.30
     and
    1.30
    Activations Density 0.083%

    No Known Activations