© 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. 1520
    Prev
    Next
    INDEX
    Explanations

    choices

    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
    orte
    -0.08
    ledged
    -0.08
    vd
    -0.08
    åĩ«
    -0.08
    èĻļ
    -0.08
    entric
    -0.08
    ç¾İæĻ¯
    -0.08
    è¨
    -0.08
    isin
    -0.08
    irty
    -0.07
    POSITIVE LOGITS
    çļĦéĢīæĭ©
    0.18
     choice
    0.17
     choices
    0.12
    choice
    0.12
     Choice
    0.12
    (choice
    0.12
    åĨ³å®ļäºĨ
    0.11
    æĺ¯åIJ¦æľī
    0.11
     choix
    0.11
    æľī没æľī
    0.11
    Activations Density 0.329%

    No Known Activations