© 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. 61077
    Prev
    Next
    INDEX
    Explanations

    most out of

    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
    Îĸ
    -0.10
     "()
    -0.09
    åħħå½ĵ
    -0.09
    é¢ĦçķĻ
    -0.09
    åIJĦæľī
    -0.09
     Pratt
    -0.08
     outlook
    -0.08
    ä¸įçķĻ
    -0.08
    irectional
    -0.08
    ä»İæł¹æľ¬
    -0.08
    POSITIVE LOGITS
    wd
    0.09
    æĪIJ绩
    0.09
    èķ´
    0.08
    èĢķ
    0.08
    æĦŁæĥħ
    0.08
     ÑĥÑģп
    0.08
    æĿ¥åģļ
    0.08
     Schn
    0.08
    _defs
    0.08
    åĩºæĿ¥äºĨ
    0.08
    Activations Density 0.080%

    No Known Activations