© 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. 34095
    Prev
    Next
    INDEX
    Explanations

    debt, assume, default, conclude

    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
     dow
    -0.08
     prow
    -0.08
    gro
    -0.08
    æ°¢èĥ½
    -0.08
     Gord
    -0.08
    顺åĪ©å®ĮæĪIJ
    -0.08
    å§Ķ书记
    -0.08
     gro
    -0.07
     indent
    -0.07
    æĺ¯ä»İ
    -0.07
    POSITIVE LOGITS
    é»ĺ认
    0.11
    失åİ»äºĨ
    0.10
    失åİ»
    0.10
    è±³
    0.09
     ÑģÑĩиÑĤ
    0.09
    _Default
    0.09
    缺å¸Ń
    0.09
    èĺħ
    0.09
    è¤Ĵ
    0.09
    afil
    0.09
    Activations Density 0.060%

    No Known Activations