© 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. 39419
    Prev
    Next
    INDEX
    Explanations

    Drop related terms

    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
    åģ»
    -0.09
    .setFill
    -0.09
    кли
    -0.09
     поиÑģк
    -0.09
    çIJ°
    -0.09
    夸
    -0.08
    èĩªåıij
    -0.08
    è¢ĭ
    -0.08
    hotel
    -0.08
    åĺ¶
    -0.08
    POSITIVE LOGITS
    .Drop
    0.12
    drops
    0.10
    -drop
    0.10
     drop
    0.09
    drop
    0.09
    _front
    0.09
     Drop
    0.09
    é¦Ķ
    0.09
    pong
    0.08
    _DROP
    0.08
    Activations Density 0.020%

    No Known Activations