© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 9739
    Prev
    Next
    INDEX
    Explanations

    build from

    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
    çľ¼çļ®
    -0.10
    æĻĹ
    -0.10
     táºŃn
    -0.09
    第åħ«
    -0.09
    åºĬä¸Ĭ
    -0.09
    å®Įç¾İ
    -0.08
    éĹ©
    -0.08
    å®Įç¾İçļĦ
    -0.08
    åĺ´éĩĮ
    -0.08
    ä»»ä½ķæŶåĢĻ
    -0.08
    POSITIVE LOGITS
    -Con
    0.10
    å¹³åĩ¡
    0.10
    isle
    0.09
    زار
    0.08
    ç§įç§į
    0.08
     available
    0.08
    ç»ıéªĮåĴĮ
    0.08
    åIJ¦
    0.08
     material
    0.08
    cai
    0.08
    Activations Density 0.157%

    No Known Activations