© 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. 22456
    Prev
    Next
    INDEX
    Explanations

    watch Kit, watch Europe, watch task

    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
    ities
    -0.12
    stras
    -0.10
    urus
    -0.09
    éĺ³åİ¿
    -0.09
    dest
    -0.08
    ality
    -0.08
    isions
    -0.08
    æ¯Ľ
    -0.08
    ads
    -0.08
    sov
    -0.08
    POSITIVE LOGITS
    able
    0.10
    amac
    0.10
    aroo
    0.10
    dog
    0.10
    tower
    0.09
     اÙĦدÙĬÙĨ
    0.09
    åĩĢåľŁ
    0.09
    paces
    0.09
    ers
    0.09
    ering
    0.09
    Activations Density 0.020%

    No Known Activations