© 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. 2434
    Prev
    Next
    INDEX
    Explanations

    professional writing and events

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

    The neuron detects source‐attribution phrases in news text—e.g. “said in a press release,” “wrote in a post,” or “explained” when attributing statements to someone.

    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
     lengkap
    0.61
     komplett
    0.57
     किल्ल्या
    0.56
    ichage
    0.55
     ग्रहों
    0.54
     propor
    0.52
     dibawah
    0.52
     cedar
    0.52
     qty
    0.52
     nè
    0.51
    POSITIVE LOGITS
     interviews
    1.36
     интервью
    1.34
     interview
    1.33
     article
    1.27
     entrevistas
    1.19
     artigo
    1.18
    インタビュー
    1.18
    interview
    1.16
     entrevista
    1.13
     статье
    1.13
    Activations Density 0.367%

    No Known Activations