© 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. 1591
    Prev
    Next
    INDEX
    Explanations

    away from or act from

    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
    ä¸Ģæŀļ
    -0.09
    追赶
    -0.09
    订
    -0.08
    éĢIJæŃ¥
    -0.08
    åĩłçĤ¹
    -0.08
    å¥ĭè¿Ľ
    -0.08
    lient
    -0.08
    èĮĨ
    -0.08
    çĽĺ
    -0.08
    æħ¢æħ¢çļĦ
    -0.08
    POSITIVE LOGITS
    è¿ľç¦»
    0.15
     away
    0.14
     khá»ıi
    0.14
    away
    0.13
     Away
    0.12
    Away
    0.11
    离å¼Ģ
    0.11
    éģ¿åħį
    0.11
    éģ¿å¼Ģ
    0.10
    éĽ¢éĸĭ
    0.10
    Activations Density 0.045%

    No Known Activations