© 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. 62864
    Prev
    Next
    INDEX
    Explanations

    pair or pairs

    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
    uario
    -0.10
    rid
    -0.09
    è¯ĦåĪĨ
    -0.08
    ridor
    -0.08
    uzz
    -0.08
    æĭŁ
    -0.08
    weg
    -0.08
    رع
    -0.08
     unn
    -0.08
    è¤Ĭ
    -0.08
    POSITIVE LOGITS
     pair
    0.26
    ä¸Ģ对
    0.25
     pairs
    0.24
    夫å¦ĩ
    0.23
    夫妻
    0.21
    _pair
    0.20
    pairs
    0.20
    pair
    0.20
     Pair
    0.20
    Pair
    0.19
    Activations Density 0.168%

    No Known Activations