© 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. 5077
    Prev
    Next
    INDEX
    Explanations

    pronoun + auxiliary verb

    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
    éĥ½æĥ³
    -0.11
    or
    -0.09
    äºĨ解ä¸Ģä¸ĭ
    -0.09
    ä¸įåı¯ç¼ºå°ij
    -0.09
    æijģ
    -0.08
    abad
    -0.08
    éĺIJæĺİ
    -0.08
    æĥ°
    -0.08
    å®ļ
    -0.08
    åħ³ä¹İ
    -0.08
    POSITIVE LOGITS
    ваÑĤ
    0.10
    éĤ£è¾¹
    0.09
    éļı
    0.09
    åĢ¾
    0.09
    _DECLARE
    0.09
    andom
    0.09
    Åĵuvre
    0.09
     Oswald
    0.08
     camper
    0.08
    çĥ¦
    0.08
    Activations Density 0.296%

    No Known Activations