© 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. 11641
    Prev
    Next
    INDEX
    Explanations

    guide you

    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
    idential
    -0.09
     minded
    -0.09
    bia
    -0.09
    æİī
    -0.08
     бÑĭ
    -0.08
    -minded
    -0.08
    -inline
    -0.08
    强è¡Į
    -0.08
    idget
    -0.08
    chia
    -0.08
    POSITIVE LOGITS
    æĮĩå¼ķ
    0.14
    (guid
    0.12
    .NewGuid
    0.12
    -guid
    0.12
     guided
    0.11
    guide
    0.11
    guid
    0.11
     guides
    0.11
     guide
    0.11
    guided
    0.11
    Activations Density 0.051%

    No Known Activations