© 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. 34-GEMMASCOPE-RES-131K
    7. 77987
    Prev
    Next
    INDEX
    Explanations

    pain relief

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

    ancient and historical names, places, and cultural references in academic and informational text.

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

    The neuron strongly activates on multi‐digit numeric tokens (especially years or large 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_34/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
    atterson
    -0.87
    ricus
    -0.81
     Мор
    -0.80
     Cao
    -0.79
    ܗ
    -0.78
     vetro
    -0.77
    osto
    -0.74
    ITION
    -0.74
     aggregation
    -0.74
    olisian
    -0.72
    POSITIVE LOGITS
     Assyrian
    1.51
     Assy
    0.89
     Senna
    0.82
     Shal
    0.77
     факт
    0.74
    ✔
    0.73
    rados
    0.73
     relieved
    0.71
    IVA
    0.71
    countries
    0.71
    Activations Density 0.009%

    No Known Activations