© 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. 5599
    Prev
    Next
    INDEX
    Explanations

    specific parameters and technical terms like 'device', 'DOMAIN', 'cost', 'Seconds', and 'Object' within code or configuration contexts

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

    This neuron strongly fires on the very first tokens after a segment or paragraph boundary—i.e. it marks the starts of new text blocks.

    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
    ود
    0.91
    ر
    0.88
    𝚞
    0.85
    un
    0.84
    indices
    0.84
    ร์
    0.82
    official
    0.82
     slags
    0.80
    ु
    0.80
    Spinner
    0.80
    POSITIVE LOGITS
     eLearning
    0.96
     Nietzsche
    0.90
     Governments
    0.89
     Anglican
    0.88
     Caesar
    0.87
     Athena
    0.86
     Neces
    0.86
     Meteor
    0.85
     Granite
    0.85
     Owl
    0.84
    Activations Density 0.000%

    No Known Activations