© 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. 18600
    Prev
    Next
    INDEX
    Explanations

    previous or current states

    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
    羣æŃ£çļĦ
    -0.09
    åħĪæĺ¯
    -0.09
    LBL
    -0.09
    æĹ©æĻļ
    -0.08
    hunt
    -0.08
    çļĦåį±å®³
    -0.08
    å®ŀåľ¨æĺ¯
    -0.08
    åħ±èµ¢
    -0.08
    羣æŃ£åģļåĪ°
    -0.08
    ç«ĭ马
    -0.08
    POSITIVE LOGITS
    å¤Ħå¢ĥ
    0.12
     own
    0.10
    ç»ıåİĨ
    0.10
    æ¿Ĥ
    0.10
     role
    0.09
    att
    0.09
    _nan
    0.09
     ÑĢолÑĮ
    0.08
     position
    0.08
    cul
    0.08
    Activations Density 0.081%

    No Known Activations