© 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. 25876
    Prev
    Next
    INDEX
    Explanations

    sports scoring and decisions

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

    The neuron fires on nouns and verbs naming specific in-game events or plays (e.g. “shootout,” “penalty,” “free throw,” “steal,” “aside”) across various sports.

    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
     unwit
    -1.06
    簸
    -1.03
     Top
    -0.98
    mère
    -0.96
     principalmente
    -0.93
     sickly
    -0.93
    才可以
    -0.93
     can
    -0.93
     altında
    -0.93
    restle
    -0.92
    POSITIVE LOGITS
    甞
    1.25
    Ⓐ
    1.19
    
    1.17
     '\\;'
    1.13
    ]()
    1.12
     decidir
    1.11
    тивы
    1.10
     decisions
    1.09
    Ⅵ
    1.08
    choices
    1.07
    Activations Density 0.002%

    No Known Activations