© 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. 54221
    Prev
    Next
    INDEX
    Explanations

    Scott Allen Perry

    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
    bose
    -0.11
    ä¸ĭéĿ¢å°±
    -0.10
    èĬ±çĵ£
    -0.10
    她说
    -0.10
    å·½
    -0.10
     ÙĨÙĤ
    -0.09
    å£ķ
    -0.09
    umph
    -0.09
    å¤ļåľ°
    -0.09
    ritis
    -0.09
    POSITIVE LOGITS
    prm
    0.10
     Belmont
    0.09
    bir
    0.09
     Mick
    0.08
    овиÑĩ
    0.08
     TOK
    0.08
    æ§
    0.08
    ibir
    0.08
    æ¡Ĥ
    0.08
     dó
    0.08
    Activations Density 0.004%

    No Known Activations