Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Gemma-3-4B-IT
    3. 22-GEMMASCOPE-2-RES-16K
    4. 409
    Prev
    Next
    INDEX
    Explanations

    This neuron strongly activates on Markdown formatting elements, particularly the `**` used for bolding and the `*` used for list items, often when they introduce new sections or bullet points in a structured text.

    oai_token-act-pair · gemini-2.5-flashTriggered by @ananthanv94
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-4b-it/resid_post/layer_22_width_16k_l0_medium
    Prompts (Dashboard)
    238,145 prompts, 512 tokens each
    Dataset (Dashboard)
    lmsys + oasst1
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     them
    0.67
     but
    0.59
     or
    0.58
     etc
    0.57
    ,
    0.54
     plus
    0.54
     me
    0.54
     ones
    0.54
     others
    0.54
     e
    0.51
    POSITIVE LOGITS
    Although
    1.45
    Despite
    1.43
    While
    1.42
    Unlike
    1.34
    During
    1.31
    The
    1.27
     Although
    1.26
    Since
    1.25
    As
    1.24
    Traditionally
    1.23
    Activations Density 5.436%

    No Known Activations