© 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. 30804
    Prev
    Next
    INDEX
    Explanations

    eduard, suryandaru, yim

    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
    ness
    -0.23
    nya
    -0.21
    ly
    -0.18
    lah
    -0.18
    na
    -0.16
    sWith
    -0.16
    ians
    -0.16
    ni
    -0.16
    ãģĦãģ¾ãģĻ
    -0.15
    mu
    -0.14
    POSITIVE LOGITS
    yla
    0.12
    ption
    0.12
    á¹ĩ
    0.12
    ptions
    0.12
    pty
    0.12
    Ùĭ
    0.11
    ÑĪка
    0.11
    gether
    0.11
    æѤåĪ»
    0.11
    pee
    0.11
    Activations Density 0.253%

    No Known Activations