© 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. 31365
    Prev
    Next
    INDEX
    Explanations

    trash

    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
    å¾Ī强
    -0.11
    彩èĻ¹
    -0.09
    ستر
    -0.09
    ä¼ļéķ¿
    -0.08
    azzi
    -0.08
     Ducks
    -0.08
    æ¹Ľ
    -0.08
    ARY
    -0.08
    SON
    -0.08
    大家
    -0.08
    POSITIVE LOGITS
    ify
    0.15
    iest
    0.13
    ification
    0.12
    iness
    0.12
    ified
    0.11
    æ»ĵ
    0.11
    ifying
    0.10
    еÑĢ
    0.10
    ingly
    0.10
    fulness
    0.10
    Activations Density 0.005%

    No Known Activations