© 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. 10-GEMMASCOPE-RES-131K
    7. 130451
    Prev
    Next
    INDEX
    Explanations

    Match followed by number

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

    The neuron fires strongly on numeric and ordinal tokens (dates, match numbers, scores, “nd,” “th,” “st,” etc.) in sports‐report contexts.

    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
     incluindo
    -1.64
    hundreds
    -1.52
    泐
    -1.47
     lámpara
    -1.43
    
    -1.42
     invitaciones
    -1.41
     musí
    -1.41
    牴
    -1.40
    蟊
    -1.40
    崀
    -1.37
    POSITIVE LOGITS
     Before
    1.84
     After
    1.75
     at
    1.60
     after
    1.57
     this
    1.48
     just
    1.46
    After
    1.46
     When
    1.45
     While
    1.44
     Unlike
    1.41
    Activations Density 0.019%

    No Known Activations