© 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. 11079
    Prev
    Next
    INDEX
    Explanations

    may or might followed by a 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
     frog
    -0.08
     ngo
    -0.08
    é¦Ļæ°´
    -0.08
    Craig
    -0.08
    ucus
    -0.08
    bud
    -0.08
    空åīį
    -0.08
    pig
    -0.08
    éĩį
    -0.08
    åij½
    -0.08
    POSITIVE LOGITS
    åįĹæµ·
    0.10
    ç¿İ
    0.09
    rex
    0.09
    éºĵ
    0.09
    æĢĢ
    0.09
    åħ³äºİæĪij们
    0.09
    çī
    0.09
    åľ¨è·¯ä¸Ĭ
    0.09
    æľ¬ç½ijç«Ļ
    0.09
     Dere
    0.09
    Activations Density 0.028%

    No Known Activations