© 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. 64895
    Prev
    Next
    INDEX
    Explanations

    breakdown by category

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

    This neuron fires on table‐style breakdown or grouping phrases—especially the “broken down by”/“by” headings used to segment data.

    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
     membres
    -1.30
    などの
    -1.25
    がある
    -1.23
    ę
    -1.15
    などを
    -1.14
    などは
    -1.13
     запись
    -1.12
    への
    -1.10
    での
    -1.10
     enorme
    -1.09
    POSITIVE LOGITS
     each
    1.66
     its
    1.47
     location
    1.34
     their
    1.34
     by
    1.34
     region
    1.31
     which
    1.28
     year
    1.26
     ethnicity
    1.24
     Each
    1.22
    Activations Density 0.027%

    No Known Activations