© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 7646
    Prev
    Next
    INDEX
    Explanations

    my purpose, guidelines, or interpretations

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

    This neuron fires on the third‐person pronouns (especially “Her,” and similar gendered pronouns) that introduce or describe a person in a biographical or profile-style passage.

    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
    人了
    0.70
    𝓐
    0.64
    quele
    0.63
    them
    0.62
    respective
    0.61
     своими
    0.60
    Zum
    0.60
     storied
    0.60
    Alors
    0.59
     اپنی
    0.59
    POSITIVE LOGITS
    目的是
    1.23
    意思是
    0.96
     내용은
    0.82
    原因是
    0.82
     είναι
    0.81
     chỉ
    0.80
     vardır
    0.79
     únicas
    0.79
     রয়েছে
    0.77
     amacı
    0.77
    Activations Density 0.383%

    No Known Activations