© 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. 32126
    Prev
    Next
    INDEX
    Explanations

    Kot, cot, or ot followed by words

    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
    èĥ¥
    -0.09
    conde
    -0.09
    atica
    -0.09
    burger
    -0.09
    inaire
    -0.08
     Fairfield
    -0.08
    æĥº
    -0.08
    å²µ
    -0.08
    initely
    -0.08
    ONSE
    -0.08
    POSITIVE LOGITS
    ting
    0.13
    tingham
    0.11
    yled
    0.10
    ÑĢеÑģ
    0.10
    ALSE
    0.10
    çĦ¶æĺ¯
    0.10
    -thirds
    0.09
    åºij
    0.09
    éĥ½å¸ĤæĬ¥
    0.09
    åŃIJéĩĮ
    0.09
    Activations Density 0.044%

    No Known Activations