© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 7326
    Prev
    Next
    INDEX
    Explanations

    IEEE 802.xx compliant

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

    The neuron strongly activates on numeric tokens—years, regulation numbers, standard identifiers, and other digit-based codes.

    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
    ⤒
    -0.75
    ليف
    -0.73
    chowy
    -0.70
    Assad
    -0.70
    MUX
    -0.69
     EVO
    -0.69
    ↤
    -0.68
    EXAMPLES
    -0.67
    ipa
    -0.67
    閱
    -0.66
    POSITIVE LOGITS
    Q
    0.97
     letter
    0.90
     KK
    0.88
     Q
    0.88
    𝐳
    0.87
     Z
    0.85
     B
    0.85
    Z
    0.83
     z
    0.82
     LETTER
    0.82
    Activations Density 0.131%

    No Known Activations