© 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. 34440
    Prev
    Next
    INDEX
    Explanations

    announced

    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
    paginator
    -0.09
    elor
    -0.09
    çľ¼çľĭ
    -0.09
    elyn
    -0.09
    lint
    -0.09
    羡æħķ
    -0.09
    Walk
    -0.08
    wer
    -0.08
    ãĤĭãĤĤãģ®
    -0.08
    iki
    -0.08
    POSITIVE LOGITS
    éĩįéĩį
    0.11
    è¿ĻåIJį
    0.09
    åij½åIJį为
    0.09
    滥
    0.09
    бÑĥд
    0.09
    å¸ĥå±Ģ
    0.09
    èµľ
    0.08
    /cgi
    0.08
    ä¼ļ使
    0.08
    æĸ°ãģŁ
    0.08
    Activations Density 0.205%

    No Known Activations