© 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. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 58677
    Prev
    Next
    INDEX
    Explanations

    words indicating speech or reporting

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    adamkarvonen/qwen3-32b-saes/saes_Qwen_Qwen3-32B_batch_top_k/resid_post_layer_32
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    éģĩä¸Ĭ
    -0.10
    ç¥ī
    -0.09
    succ
    -0.09
    åı«
    -0.08
    pard
    -0.08
    æİĤ
    -0.08
    触
    -0.08
    建ç«ĭèµ·
    -0.08
    å°±å¤Ł
    -0.08
    éĩįçİ°
    -0.08
    POSITIVE LOGITS
    çļĦåĽŀçŃĶ
    0.22
    åĽŀçŃĶ
    0.21
     answer
    0.17
    çŃĶ
    0.16
    çļĦçŃĶæ¡Ī
    0.16
    çŃĶå¤į
    0.16
     оÑĤвеÑĤ
    0.15
    çŃĶãģĪ
    0.15
     reply
    0.15
     answers
    0.15
    Activations Density 0.061%

    No Known Activations