© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 1466
    Prev
    Next
    INDEX
    Explanations

    The neuron fires on elongated, expressive words where letters are stretched (e.g. “pleeeaaassse,” “loooove,” “ooooooo”) to convey emphasis or excitement.

    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
     counterv
    0.89
     fragmentary
    0.74
     "
    0.73
     interwoven
    0.72
     corticoster
    0.71
     subsidiary
    0.71
     hitherto
    0.70
     subsp
    0.70
     adherents
    0.69
     systemat
    0.69
    POSITIVE LOGITS
    uuuu
    1.53
    eeee
    1.52
    iiii
    1.40
    ahaha
    1.38
    hhhh
    1.37
    😭😭
    1.37
    ssss
    1.36
    hhh
    1.34
    pppp
    1.34
    eee
    1.32
    Activations Density 0.390%

    No Known Activations