© 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. 18976
    Prev
    Next
    INDEX
    Explanations

    numbers followed by percent or time

    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
    иÑĢован
    -0.08
    atform
    -0.08
    ä¹Łæ¯Ķè¾ĥ
    -0.08
    åıĪ好
    -0.08
    è´¨
    -0.07
    æĹ¢çĦ¶
    -0.07
    å¤Ķ
    -0.07
    (Display
    -0.07
    ouro
    -0.07
    ä¹°åĪ°
    -0.07
    POSITIVE LOGITS
     ones
    0.14
    æģ°æģ°
    0.11
    常æĢģ
    0.10
    çĭłæĬĵ
    0.10
    è¿Ļæł·åģļ
    0.10
    æķ´æķ´
    0.10
    常è§Ħ
    0.09
     Ones
    0.09
    ä¹ĭåīįçļĦ
    0.09
    phans
    0.09
    Activations Density 0.118%

    No Known Activations