© 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. 58110
    Prev
    Next
    INDEX
    Explanations

    assuming followed by a condition

    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
    èĻ½
    -0.11
    èĻ½çĦ¶
    -0.09
    èĻ½çĦ¶æĺ¯
    -0.08
    al
    -0.08
    gence
    -0.08
    磼
    -0.08
    éĽĸ
    -0.08
    çĿĢçľ¼äºİ
    -0.08
    å°½
    -0.08
    åħįç¨İ
    -0.08
    POSITIVE LOGITS
    eer
    0.10
    æĪijåİ»
    0.10
    顺åĪ©
    0.09
    ä½łæĥ³
    0.09
    é£İäºij
    0.09
    tons
    0.09
    ering
    0.09
    bst
    0.09
    assuming
    0.09
    æĤ¨
    0.09
    Activations Density 0.023%

    No Known Activations