© 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
    2. Gemma-3-12B
    3. 24-GEMMASCOPE-2-RES-16K
    4. 426
    Prev
    Next
    INDEX
    Explanations

    The neuron is essentially a “start-of‐sequence” detector, firing strongly on the special BOS token.

    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
    。
    0.29
    Pokemon
    0.28
    とか
    0.27
     и
    0.27
     কিংবা
    0.27
    といった
    0.26
    ☠
    0.26
    Combat
    0.26
     ্
    0.26
    なども
    0.25
    POSITIVE LOGITS
     యొక్క
    0.34
    本身的
    0.30
     مذکور
    0.29
     undergoes
    0.29
    ஸின்
    0.29
     intends
    0.27
    টিকে
    0.27
     receives
    0.27
     possesses
    0.27
    內的
    0.27
    Activations Density 16.600%

    No Known Activations