© 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. 55937
    Prev
    Next
    INDEX
    Explanations

    from time to time

    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
    å°ıæŶåīį
    -0.10
    ç»ĨåĮĸ
    -0.09
     Walsh
    -0.08
    æŁ¿
    -0.08
    uden
    -0.08
    avit
    -0.08
     hasNext
    -0.08
     бÑĥкв
    -0.08
    åĵº
    -0.08
    lector
    -0.08
    POSITIVE LOGITS
    åģ¶å°Ķ
    0.22
     occasional
    0.19
     occasionally
    0.19
    æŶä¸įæŶ
    0.17
     Occasionally
    0.16
    åģ¶
    0.15
    æľīæŶ
    0.14
     иногда
    0.12
     ocas
    0.12
    æľīæŶåĢĻ
    0.12
    Activations Density 0.042%

    No Known Activations