© 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. 24-GEMMASCOPE-2-RES-16K
    7. 11100
    Prev
    Next
    INDEX
    Explanations

    proposed strategies

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

    The neuron fires strongly on structural or formatting markers—things like standalone dashes, ellipses, parentheses, and “available” in amenity‐type listings—i.e. non-prose tokens that mark section breaks or side/detail information.

    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
    ifax
    0.88
    als
    0.86
    Just
    0.85
     ทำให้
    0.82
    Now
    0.78
    Today
    0.77
    Try
    0.77
    erende
    0.76
    Of
    0.76
    Until
    0.75
    POSITIVE LOGITS
     grievance
    0.82
    摂取
    0.79
     cursory
    0.77
     dior
    0.76
     peritoneal
    0.76
    摄
    0.76
     grip
    0.75
     foodservice
    0.74
     scoop
    0.73
     skimming
    0.73
    Activations Density 0.004%

    No Known Activations