© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 73567
    Prev
    Next
    INDEX
    Explanations

    going through

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

    The neuron activates on content-bearing words (e.g. nouns, verbs, and other non–stopword tokens) rather than on common function 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
     looming
    -0.98
     lost
    -0.96
     скла
    -0.94
     follow
    -0.93
    FORCEMENT
    -0.91
    在這
    -0.90
    eafter
    -0.89
     soil
    -0.88
     foolish
    -0.88
     Synag
    -0.87
    POSITIVE LOGITS
     into
    1.16
     through
    1.12
     hindurch
    1.09
     attraverso
    1.04
     niego
    0.96
     to
    0.94
     através
    0.93
    pog
    0.91
    čních
    0.91
    ⁃
    0.91
    Activations Density 0.073%

    No Known Activations