© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 27073
    Prev
    Next
    INDEX
    Explanations

    CAS numbers and sequences

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

    programming and mathematical notation involving colons, parentheses, and other special characters in coding contexts.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    This neuron detects text describing items being “picked without replacement from {…},” i.e. combinatorial probability setups.

    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
     any
    -1.20
     additional
    -1.18
     if
    -1.09
     always
    -1.02
     had
    -1.00
     seem
    -1.00
    〜。
    -0.98
     some
    -0.97
     two
    -0.95
     new
    -0.94
    POSITIVE LOGITS
     lunares
    1.26
     dépens
    1.17
     siè
    1.16
     longues
    1.15
     renou
    1.13
     clés
    1.13
     corriger
    1.13
     triomphe
    1.13
     logiciels
    1.12
     travailler
    1.10
    Activations Density 0.008%

    No Known Activations