© 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. 52959
    Prev
    Next
    INDEX
    Explanations

    `wh` questions followed by pronouns

    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
    è¡Įäºĭ
    -0.09
    reat
    -0.08
    äºĭä¸ļåıijå±ķ
    -0.08
    å¨Ħ
    -0.08
    oard
    -0.08
     apt
    -0.07
    ä½ľä¸ºä¸Ģç§į
    -0.07
    èĿĪ
    -0.07
    oo
    -0.07
    typeof
    -0.07
    POSITIVE LOGITS
    otime
    0.10
    ling
    0.10
    说äºĨ
    0.10
    æħİ
    0.09
    è§Ħå®ļçļĦ
    0.09
    azzi
    0.09
    è³ĵ
    0.09
    LING
    0.09
    宾
    0.09
    unate
    0.09
    Activations Density 0.009%

    No Known Activations