© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 73982
    Prev
    Next
    INDEX
    Explanations

    versus, against, vs

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

    The neuron flags head-to-head/versus expressions (e.g. “7 v 7,” “1 v 1,” “one-on-one”) in sports‐style or competitive 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
     
    -1.72
    ):
    -1.69
    dä
    -1.58
    adır
    -1.53
    ),
    -1.50
    ?
    -1.48
    ATION
    -1.48
     município
    -1.47
    远处
    -1.45
     móvel
    -1.45
    POSITIVE LOGITS
    s
    1.80
     spetta
    1.70
     they
    1.60
    凇
    1.48
    λα
    1.43
     квартира
    1.42
     pittore
    1.41
    ims
    1.38
    this
    1.38
    elance
    1.38
    Activations Density 0.006%

    No Known Activations