© 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. 12603
    Prev
    Next
    INDEX
    Explanations

    contrast and softening

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

    The neuron activates strongly on technical discussion of colors, hues, and color mixing in text.

    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
    滓
    0.60
    娯
    0.57
    服用
    0.54
    甛
    0.54
    አ
    0.54
    芸術
    0.52
     香港
    0.52
    詳細
    0.51
    Descripción
    0.50
     фильмы
    0.50
    POSITIVE LOGITS
     contrast
    1.19
     contrasting
    1.13
     contrasts
    1.10
     Contrast
    1.10
     juxtaposition
    1.07
    Contrast
    1.03
     kontrast
    1.01
     contrasted
    0.99
    contrast
    0.98
     Contr
    0.98
    Activations Density 0.705%

    No Known Activations