© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 3090
    Prev
    Next
    INDEX
    Explanations

    that aligns

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

    The neuron specifically fires on occurrences of the word “that” when it’s used as a relative pronoun or conjunction introducing a subordinate clause.

    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
    都可以
    1.09
    are
    1.07
    都要
    1.07
     являются
    0.97
    均
    0.96
    都需要
    0.95
     incidentally
    0.94
    都
    0.94
    都被
    0.93
    ている
    0.93
    POSITIVE LOGITS
     وأن
    1.25
     iż
    0.95
     stara
    0.94
    Ꮹ
    0.93
     pessoa
    0.91
     Wasn
    0.91
     masalah
    0.90
     zwią
    0.90
     atrás
    0.89
    ल
    0.88
    Activations Density 0.294%

    No Known Activations