© 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. 878
    Prev
    Next
    INDEX
    Explanations

    The neuron is tuned to high-frequency function words (auxiliaries, pronouns, inflectional particles) that often occur in instructional/advisory phrasing.

    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
    čne
    0.44
    žil
    0.40
    ḥ
    0.40
    ari
    0.39
    it
    0.39
     linkages
    0.39
    ô
    0.38
    ilience
    0.38
    ukka
    0.38
    ø
    0.37
    POSITIVE LOGITS
     乀
    0.55
    Someone
    0.50
     bạn
    0.50
    讒
    0.49
    洟
    0.49
     montaña
    0.48
    Você
    0.47
    Usuario
    0.46
     pequeño
    0.46
    あなたの
    0.46
    Activations Density 0.001%

    No Known Activations