© 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. 32356
    Prev
    Next
    INDEX
    Explanations

    that or those followed by a noun

    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
    åī½
    -0.12
    æ»ģ
    -0.11
    ä»Ģä¹Īåij¢
    -0.10
    æĬĹæĭĴ
    -0.09
     POLIT
    -0.09
     Fres
    -0.09
    oure
    -0.09
    ç¢
    -0.09
    éħŀ
    -0.09
    女ç¥ŀ
    -0.09
    POSITIVE LOGITS
    UPLE
    0.09
    aways
    0.08
    omba
    0.08
    cene
    0.08
     occup
    0.08
    åħħåĪĨåĪ©çĶ¨
    0.08
    paced
    0.08
    /request
    0.08
     JT
    0.08
    forth
    0.08
    Activations Density 0.140%

    No Known Activations