© 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. 39890
    Prev
    Next
    INDEX
    Explanations

    produces outcomes

    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
    ight
    -0.10
    ÑĢеÑģ
    -0.10
    ousel
    -0.09
    ìĤ¬íķŃ
    -0.09
    ÑĢиÑģ
    -0.09
    .Roll
    -0.09
    ties
    -0.09
    ort
    -0.09
    å¿ĥæĢĿ
    -0.09
    rollers
    -0.09
    POSITIVE LOGITS
    ä¸įäºĨ
    0.14
    ively
    0.12
    ivity
    0.11
    äºĨä¸Ģ
    0.10
    äºĨä¸ĢäºĽ
    0.10
    è´ŁéĿ¢å½±åĵį
    0.10
    iland
    0.09
    äºĨä¸Ģ个
    0.09
    er
    0.08
    ably
    0.08
    Activations Density 0.068%

    No Known Activations