© 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. 39382
    Prev
    Next
    INDEX
    Explanations

    followed by punctuation or specific 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
    åĩĮ
    -0.08
    normally
    -0.08
    nal
    -0.08
    åĩłåįģå¹´
    -0.08
    åĽŀé¦Ī
    -0.08
    æŃ¦å£«
    -0.08
    ç»ĻåĬĽ
    -0.08
    HEST
    -0.08
    urette
    -0.08
    æ··æ·Ĩ
    -0.08
    POSITIVE LOGITS
    使æĪij
    0.11
    æĪijä¸Ģ缴
    0.10
    åģľçķĻ
    0.09
    æĺ¯æĪijçļĦ
    0.09
    æĪijçĽ¸ä¿¡
    0.09
     gönder
    0.08
    åĽłä¸ºæĪij
    0.08
    ä½ĨæĪij
    0.08
    æĻļé¥Ń
    0.08
    æĻ¤
    0.08
    Activations Density 0.033%

    No Known Activations