© 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. 10-GEMMASCOPE-RES-131K
    7. 71452
    Prev
    Next
    INDEX
    Explanations

    scientific terms

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

    technical terminology in academic or scientific contexts.

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

    The neuron strongly activates on numeric tokens (especially floating-point numbers).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
     and
    -2.06
     kteří
    -1.70
     helft
    -1.61
     betrek
    -1.60
    ↵↵
    -1.56
    ası
    -1.55
    </th>
    -1.48
     zijne
    -1.48
    '),
    -1.47
     minimale
    -1.47
    POSITIVE LOGITS
     it
    2.19
     there
    2.09
    スポンサーリンク
    1.89
     we
    1.84
     wollten
    1.77
     diversa
    1.76
     other
    1.75
     this
    1.74
     verbessert
    1.70
     ähn
    1.69
    Activations Density 0.000%

    No Known Activations