© 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. 31-GEMMASCOPE-2-RES-16K
    7. 330
    Prev
    Next
    INDEX
    Explanations

    The neuron is specifically responding to the start‐of‐sequence marker, i.e. it fires strongly on the “<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_31_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
    NSInteger
    0.31
     जमान
    0.30
    webhook
    0.29
    𝑚
    0.29
    groovy
    0.29
    ንዱ
    0.29
    MORDOR
    0.29
    💻
    0.29
     metaverse
    0.28
    롯데
    0.28
    POSITIVE LOGITS
     William
    0.66
     Robert
    0.63
     John
    0.62
     Joseph
    0.61
     Charles
    0.60
     Richard
    0.59
     Peter
    0.59
     David
    0.58
     George
    0.58
     Margaret
    0.57
    Activations Density 6.011%

    No Known Activations