© 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. 29903
    Prev
    Next
    INDEX
    Explanations

    i have

    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
    æĪĸå¤ļ
    -0.11
    mie
    -0.10
    è¦ģæľī
    -0.09
    aten
    -0.09
    åįłæľī
    -0.09
    åħ¶æīĢ
    -0.09
    sic
    -0.09
    ä»ĸå¦Ī
    -0.08
    æĻ°
    -0.08
    ä»Ģä¹Īåij¢
    -0.08
    POSITIVE LOGITS
    senal
    0.09
    ç¥Ŀ
    0.09
     Gingrich
    0.09
     communicator
    0.09
    zelf
    0.08
    æĬķ
    0.08
    /th
    0.08
    elman
    0.08
    /respond
    0.08
    Ñħо
    0.08
    Activations Density 0.048%

    No Known Activations