© 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. 54409
    Prev
    Next
    INDEX
    Explanations

    most others, examples, opponents

    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
    马åĬĽ
    -0.09
    马路
    -0.09
    åľ°å¸¦
    -0.09
     пеÑĢе
    -0.08
    åħ±å»º
    -0.08
    å·®çĤ¹
    -0.08
    åħĢ
    -0.08
    ported
    -0.08
    æĢ»æľī
    -0.08
    åIJĦ大
    -0.08
    POSITIVE LOGITS
    quo
    0.10
    åħ¶å®ĥ
    0.09
    owo
    0.09
    iras
    0.09
    _alive
    0.09
    ฯ
    0.08
    olvers
    0.08
    å°±æĺ¯è¦ģ
    0.08
    ilar
    0.08
    .MSG
    0.08
    Activations Density 0.017%

    No Known Activations