© 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. 33896
    Prev
    Next
    INDEX
    Explanations

    hostility, conflict, motivation, impulsiveness

    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
    没æľīä»»ä½ķ
    -0.09
     corre
    -0.09
    éªĩ
    -0.08
    educated
    -0.08
    lut
    -0.08
     adult
    -0.08
    åºĸ
    -0.08
    ÑģÑĤÑĢа
    -0.08
    æĹłæķ°
    -0.08
    å¯Ł
    -0.08
    POSITIVE LOGITS
    çĿ¥
    0.11
    å°ıå·§
    0.10
    -flex
    0.10
    æ·¦
    0.09
    åĵ®
    0.09
    游è§Ī
    0.09
    رج
    0.08
    ä¸ºæł¸å¿ĥ
    0.08
    宿
    0.08
    flex
    0.08
    Activations Density 0.480%

    No Known Activations