© 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. 3978
    Prev
    Next
    INDEX
    Explanations

    Yaqiang Wang, Jaakko Keränen, Yasuo Kiyoki, Lajos Molnár, Sandro Vukosavljevic, Aki Tzanakakis

    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
    ality
    -0.11
    ering
    -0.11
    ered
    -0.10
    ãĤį
    -0.10
    afen
    -0.09
    ened
    -0.09
    ï¸
    -0.09
    å¥Ĺé¤IJ
    -0.09
    edList
    -0.09
    elves
    -0.09
    POSITIVE LOGITS
    lı
    0.13
    cı
    0.12
    ın
    0.11
    éªĿ
    0.11
    alysis
    0.11
    sembler
    0.11
    ında
    0.11
     пÑĢовед
    0.11
    ðŁı»
    0.11
    ience
    0.10
    Activations Density 1.990%

    No Known Activations