© 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. 21706
    Prev
    Next
    INDEX
    Explanations

    album

    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
    icio
    -0.09
    flux
    -0.09
    iez
    -0.09
    reek
    -0.09
    Reddit
    -0.09
    女åŃ©
    -0.08
    第ä¸Ģæī¹
    -0.08
     prec
    -0.08
    å¼ĥ
    -0.08
     girl
    -0.08
    POSITIVE LOGITS
    åħ»èĢģ
    0.13
    èĢģå¹´
    0.12
    èĢģ头
    0.12
    éĢĢä¼ij
    0.11
    æĻļå¹´
    0.11
    æłĩå¿ĹæĢ§
    0.11
    èĸľ
    0.11
    è¡°èĢģ
    0.11
    èĢĦ
    0.11
     pension
    0.11
    Activations Density 0.165%

    No Known Activations