© 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. 35030
    Prev
    Next
    INDEX
    Explanations

    approval and disapproval

    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
    ally
    -0.11
    aways
    -0.11
    eros
    -0.11
     Ñģобой
    -0.10
    eker
    -0.10
    ãģıãĤĭ
    -0.10
    ry
    -0.10
    wear
    -0.09
    aro
    -0.09
    à¹Į
    -0.09
    POSITIVE LOGITS
    ably
    0.15
    ment
    0.13
    /dis
    0.11
    ä¸įäºĨ
    0.10
    uste
    0.10
    ãĥ¡ãĥ³ãĥĪ
    0.10
    رÙĥ
    0.09
    人士
    0.09
    ptions
    0.09
    uable
    0.09
    Activations Density 0.013%

    No Known Activations