© 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. 64352
    Prev
    Next
    INDEX
    Explanations

    eclipse, deadline

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

    The neuron consistently fires on numeric tokens, i.e. sequences of digits representing values (counts, scales, versions, etc.).

    oai_token-act-pair · o4-mini
    Triggered 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
    Negative Logits
     their
    -1.27
     because
    -1.15
     even
    -1.15
     to
    -1.13
     for
    -1.12
     or
    -1.04
     more
    -1.04
     each
    -1.02
     some
    -1.02
     when
    -0.99
    POSITIVE LOGITS
     défauts
    1.21
     thua
    1.14
    を作って
    1.10
    ⮑
    1.09
     nạ
    1.09
     bì
    1.05
    significantly
    1.05
     permettant
    1.05
    essentially
    1.04
    effectively
    1.04
    Activations Density 0.080%

    No Known Activations