© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 21338
    Prev
    Next
    INDEX
    Explanations

    comic titles and characters

    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
     Revenge
    -0.09
    æĢģ
    -0.09
    forward
    -0.09
    æī§æĶ¿
    -0.08
    isel
    -0.08
     Stateless
    -0.08
    aravel
    -0.08
     Stra
    -0.08
     Tween
    -0.08
    bet
    -0.08
    POSITIVE LOGITS
    èĢĹ
    0.10
    åĸ§
    0.10
    æĸ©
    0.10
    åļ£
    0.10
    ductor
    0.09
     оÑĤд
    0.09
    duit
    0.09
    çĨŁæĤīçļĦ
    0.09
    æľīä¿¡å¿ĥ
    0.09
    ç£ģåľº
    0.09
    Activations Density 0.049%

    No Known Activations