© 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. 27893
    Prev
    Next
    INDEX
    Explanations

    express

    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
    ÑĪки
    -0.10
    èµĸ以çĶŁåŃĺ
    -0.09
    clock
    -0.09
    ä¸Ģåij³
    -0.09
    亲åıĭ
    -0.09
    çľĭå¾Ĺ
    -0.08
    ep
    -0.08
    REN
    -0.08
    orest
    -0.08
    acha
    -0.08
    POSITIVE LOGITS
    iveness
    0.14
    ively
    0.14
    ivity
    0.12
    ions
    0.11
    sing
    0.10
    ãģĭãģ¤
    0.10
    ibility
    0.09
    ÙħÙĬÙĨ
    0.09
    ingly
    0.09
    èĩ¾
    0.09
    Activations Density 0.020%

    No Known Activations