© 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. 58
    Prev
    Next
    INDEX
    Explanations

    The neuron strongly responds to sentence-beginning “This” tokens, i.e. it detects when a new sentence starts with the word “This.”

    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
    E
    0.30
    ・
    0.30
    、
    0.30
     Mỗi
    0.29
     カ
    0.28
    0
    0.28
    AL
    0.28
     Alguns
    0.28
     ค
    0.28
    AR
    0.28
    POSITIVE LOGITS
     particular
    0.53
     newfound
    0.49
     phenomenon
    0.47
     momentous
    0.46
    particular
    0.46
     idyllic
    0.44
     particolare
    0.44
     same
    0.43
     latter
    0.43
     article
    0.43
    Activations Density 0.916%

    No Known Activations