© 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. 10-GEMMASCOPE-RES-131K
    7. 86051
    Prev
    Next
    INDEX
    Explanations

    entry

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

    The neuron strongly activates on numeric tokens—digits and number-like strings (e.g. IDs, measurements, percentages).

    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
    C
    -2.77
    -
    -2.34
    H
    -2.31
    O
    -2.22
            
    -2.20
    and
    -2.16
    衕
    -2.14
    و
    -2.11
    en
    -2.09
    OS
    -1.99
    POSITIVE LOGITS
     besonderes
    2.41
    砹
    2.38
     imprimi
    2.36
     ganzes
    2.34
     brinda
    2.27
     入っ
    2.22
     शुर
    2.13
     verbess
    2.13
    贅
    2.11
    sudadera
    2.09
    Activations Density 0.020%

    No Known Activations