© 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. 24-GEMMASCOPE-2-RES-16K
    7. 13865
    Prev
    Next
    INDEX
    Explanations

    another, observations, further

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

    The neuron strongly activates on discourse-level connectives and transition words (e.g. and, but, then, so, if, to) that structure the flow of exposition.

    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
    en
    0.66
    .
    0.64
    ↵
    0.63
    utilizzo
    0.60
     सिर्फ
    0.59
    Whats
    0.59
    í
    0.59
    只需
    0.58
    ри
    0.58
    ways
    0.58
    POSITIVE LOGITS
     another
    0.88
     Another
    0.77
     ANOTHER
    0.75
     otro
    0.68
     weiteres
    0.67
     থাকিতে
    0.66
     AcOEt
    0.66
    拡散方程式
    0.66
    再び
    0.66
     weitere
    0.65
    Activations Density 15.677%

    No Known Activations