© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 10490
    Prev
    Next
    INDEX
    Explanations

    shadows and 'fall'

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

    The neuron is activated by occurrences of the word “shadow.”

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    the word "shadow" or phrases describing shadows and light-related spatial concepts.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @jyhe0408

    references to shadows and light, especially in contexts of illumination, eclipses, and optical effects.

    oai_token-act-pair · gpt-5Triggered 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
    rophic
    0.61
     Fireworks
    0.59
    bricks
    0.56
    Fireworks
    0.56
    语音
    0.55
     disassemble
    0.55
     பார்
    0.53
    粨
    0.53
     উপল
    0.52
     firepower
    0.52
    POSITIVE LOGITS
     shadow
    2.61
     shadows
    2.41
    Shadow
    2.38
     Shadow
    2.38
    shadow
    2.36
    影子
    2.23
     Shadows
    2.14
     sombra
    2.05
     sombras
    2.03
     shadowing
    1.99
    Activations Density 0.125%

    No Known Activations