© 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. 56994
    Prev
    Next
    INDEX
    Explanations

    canyon, gorge, or French

    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
    ewan
    -0.09
    ANEL
    -0.09
    vious
    -0.09
    à¹ĩ
    -0.08
    ieval
    -0.08
    -ups
    -0.08
    idious
    -0.08
    etable
    -0.08
    uth
    -0.08
    acre
    -0.08
    POSITIVE LOGITS
    æ¿ij
    0.11
    eer
    0.10
    è®¤çľŁåŃ¦ä¹ł
    0.10
    å£ij
    0.09
    orm
    0.09
    ones
    0.09
    éĩĮçļĦ
    0.09
    ously
    0.09
    æ·±å¤Ħ
    0.09
    lauf
    0.09
    Activations Density 0.008%

    No Known Activations