© 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. 26665
    Prev
    Next
    INDEX
    Explanations

    vast amount

    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
    thal
    -0.09
    athers
    -0.09
    tera
    -0.08
    wards
    -0.08
    ONT
    -0.08
    trip
    -0.08
    ÑĤÑĢ
    -0.08
    oine
    -0.08
    than
    -0.08
    ond
    -0.08
    POSITIVE LOGITS
    ively
    0.14
    ness
    0.14
    acre
    0.12
    å¼ĢæĿ¥
    0.12
    iveness
    0.12
     amounts
    0.11
    ening
    0.11
    sword
    0.11
     majority
    0.10
    NESS
    0.10
    Activations Density 0.008%

    No Known Activations