© 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. 6328
    Prev
    Next
    INDEX
    Explanations

    contrastive conjunctions

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

    The neuron fires on contrastive discourse markers (e.g. “however”).

    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
    ных
    0.88
    ി
    0.86
     ampliar
    0.86
    ו
    0.84
    ி
    0.84
     inoltre
    0.80
     asimismo
    0.80
     ervoor
    0.79
     επίσης
    0.79
     使っ
    0.79
    POSITIVE LOGITS
    并不是
    0.88
    并不
    0.82
    更
    0.79
    데
    0.77
    ent
    0.77
    quite
    0.76
    是不
    0.74
    是有
    0.73
    thy
    0.73
    并非
    0.73
    Activations Density 0.057%

    No Known Activations