© 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. 5895
    Prev
    Next
    INDEX
    Explanations

    thoroughly described or investigated

    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
    æİĩ
    -0.09
     Gef
    -0.09
    éļIJéļIJ
    -0.08
    å½·
    -0.08
    nable
    -0.08
    ife
    -0.08
    folk
    -0.08
    avier
    -0.08
     hend
    -0.08
    æģį
    -0.08
    POSITIVE LOGITS
    å½»åºķ
    0.12
    ties
    0.12
     thoroughly
    0.11
    going
    0.11
    bred
    0.11
    orough
    0.10
    äºļ马
    0.10
    å¾¹
    0.10
    å½»
    0.10
    neck
    0.10
    Activations Density 0.010%

    No Known Activations