© 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. 11697
    Prev
    Next
    INDEX
    Explanations

    voting results and decisions

    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
    spell
    -0.08
    豪
    -0.08
    æ¯ı次
    -0.08
    nc
    -0.08
    åı©
    -0.08
     spell
    -0.08
    Cheap
    -0.08
     spilled
    -0.08
    ä¸įä¿¡
    -0.08
     Cheap
    -0.07
    POSITIVE LOGITS
    表åĨ³
    0.16
    æĬķ票
    0.15
     vote
    0.13
     unanimously
    0.13
     voting
    0.12
     votes
    0.11
    vote
    0.11
     голоÑģ
    0.11
    Vote
    0.10
     Vote
    0.10
    Activations Density 0.047%

    No Known Activations