© 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-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 6758
    Prev
    Next
    INDEX
    Explanations

    red, redness, crimson

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

    The neuron fires on technology‐related product or project names—particularly “Ruby” and “Red*” software names (e.g. Ruby on Rails, Redmine, Red Hat).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-12b-pt/resid_post/layer_31_width_16k_l0_medium
    Prompts (Dashboard)
    392,802 prompts, 256 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     مقامی
    0.60
     gör
    0.59
     inet
    0.59
     ახ
    0.57
    畯
    0.57
     ამ
    0.57
     jiwa
    0.57
     shave
    0.55
     grey
    0.54
     LORD
    0.53
    POSITIVE LOGITS
    Red
    1.75
     Red
    1.69
     RED
    1.44
     红
    1.39
    getRed
    1.38
    紅
    1.36
    红
    1.35
    RED
    1.33
     레드
    1.30
     red
    1.28
    Activations Density 0.192%

    No Known Activations