© 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. 54317
    Prev
    Next
    INDEX
    Explanations

    twisting

    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
    åĹij
    -0.10
    rou
    -0.10
    hots
    -0.10
    iance
    -0.09
    éĢ¸
    -0.09
    çĦIJ
    -0.09
    峪
    -0.09
    naissance
    -0.09
    å¾Ħ
    -0.09
    ëĭ´
    -0.08
    POSITIVE LOGITS
    èĻ¬
    0.11
    以èī²
    0.10
    å¤ļ彩
    0.10
    ENSION
    0.09
    æĥķ
    0.09
    ä»ģ
    0.09
    ention
    0.08
    ðŁı»
    0.08
    鼩
    0.08
    ä¸ºè¿Ľä¸ĢæŃ¥
    0.08
    Activations Density 0.013%

    No Known Activations