© 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. 22-GEMMASCOPE-RES-131K
    7. 26429
    Prev
    Next
    INDEX
    Explanations

    times and places

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

    The neuron activates on numeric tokens—especially measurements and decimal values—marking where the text gives precise numbers or angles.

    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
    หรือ
    -1.29
     validar
    -1.27
    ilustra
    -1.17
     saltar
    -1.16
     despertar
    -1.15
     sorprender
    -1.14
     multiuso
    -1.11
     inserir
    -1.11
     sostener
    -1.11
    entuk
    -1.11
    POSITIVE LOGITS
    imetry
    1.24
    −/−
    1.20
    oted
    1.20
    ></
    1.20
     usw
    1.19
    itively
    1.18
    ‧
    1.17
     («
    1.13
    ceae
    1.13
     /*
    1.13
    Activations Density 0.084%

    No Known Activations