© 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. 24-GEMMASCOPE-2-RES-16K
    7. 5608
    Prev
    Next
    INDEX
    Explanations

    dialogue prompt examples

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

    The neuron strongly activates on tokens that begin new paragraphs (i.e. tokens appearing immediately after blank‐line or document boundaries).

    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_24_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
    बाग
    0.67
    (!)
    0.64
    k
    0.64
    跻
    0.60
    b
    0.59
     (!)
    0.58
    g
    0.58
    Así
    0.56
    den
    0.55
     ­
    0.55
    POSITIVE LOGITS
     nums
    0.75
     FontSize
    0.68
     vertices
    0.68
     plugins
    0.66
     inodes
    0.66
     함수
    0.65
     scanf
    0.64
     embeddings
    0.64
     phd
    0.64
     theorems
    0.63
    Activations Density 3.345%

    No Known Activations