© 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. 34-GEMMASCOPE-RES-131K
    7. 78558
    Prev
    Next
    INDEX
    Explanations

    quarterback actions and performance

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

    The main thing this neuron does is identify numeric tokens—digits and statistical numbers in sports text.

    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
    GetKey
    -0.81
     כח
    -0.80
    ΠΑ
    -0.79
    IConfiguration
    -0.79
     defensive
    -0.79
     violin
    -0.79
     Gson
    -0.78
    tack
    -0.77
    陂
    -0.77
    riceps
    -0.77
    POSITIVE LOGITS
     throwing
    1.55
    throwing
    1.38
     throws
    1.32
    QB
    1.29
     QB
    1.29
     Throwing
    1.26
    Throwing
    1.23
     threw
    1.16
     quarterback
    1.16
     accuracy
    1.10
    Activations Density 0.039%

    No Known Activations