© 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. 53925
    Prev
    Next
    INDEX
    Explanations

    and followed by names/titles

    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
    olum
    -0.08
    å§Ķ
    -0.08
    ãģĹãĤĩãģĨ
    -0.07
    å±¥
    -0.07
     path
    -0.07
    ulan
    -0.07
    åĢı
    -0.07
     all
    -0.07
    Them
    -0.07
     meaning
    -0.07
    POSITIVE LOGITS
    ä¸ĢäºĽ
    0.14
    ients
    0.13
    ä¸Ģç³»åĪĹ
    0.12
    èĭ¥å¹²
    0.12
    æīĭä¸ŃçļĦ
    0.11
    ä½Ĩä¸įéĻIJäºİ
    0.10
    éĤ£äºĽ
    0.10
    ä¸Ģ个个
    0.10
    åħ¶å®ĥ
    0.10
    äºī缸
    0.10
    Activations Density 0.386%

    No Known Activations