© 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. 4059
    Prev
    Next
    INDEX
    Explanations

    half split

    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
    åİĮ
    -0.10
     конкÑĢ
    -0.10
    á»ĩt
    -0.09
    çĽĨ
    -0.09
    elon
    -0.09
    agraph
    -0.09
    .Deserialize
    -0.09
    cone
    -0.09
    æŶä¸į
    -0.09
    .ping
    -0.08
    POSITIVE LOGITS
     halves
    0.10
    æĭĪ
    0.09
     Hal
    0.09
    尽头
    0.09
     мн
    0.09
    èĩ¾
    0.08
    ä¸ĢåįĬ
    0.08
     hal
    0.08
    hal
    0.08
    7
    0.08
    Activations Density 0.021%

    No Known Activations