© 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. 32925
    Prev
    Next
    INDEX
    Explanations

    previous words, or endings

    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
    coat
    -0.09
    éĢĥ离
    -0.08
    éķĸ
    -0.08
    éĴ¥
    -0.08
    ople
    -0.08
    éĢĥ
    -0.08
    åįģå¹´åīį
    -0.08
    Bon
    -0.07
    åĬĿ
    -0.07
    ouse
    -0.07
    POSITIVE LOGITS
    管çIJĨå·¥ä½ľ
    0.09
    访
    0.09
    触åıĬ
    0.09
    isted
    0.09
    çļĦ表çİ°
    0.09
    æļĸæ°Ķ
    0.08
    åij½ä¸Ń
    0.08
    ä¿İ
    0.08
    åĩĨå¤ĩå·¥ä½ľ
    0.08
    å¤©åľ°
    0.08
    Activations Density 0.188%

    No Known Activations