© 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. 48631
    Prev
    Next
    INDEX
    Explanations

    Ariel, Avery, Fletcher

    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
    ards
    -0.11
    /her
    -0.10
    stead
    -0.10
    å¾Ĺèµ·
    -0.10
    ðŁı»
    -0.09
    olo
    -0.09
    ä¹İ
    -0.09
    uch
    -0.09
    isl
    -0.09
    ertext
    -0.09
    POSITIVE LOGITS
    ../../../
    0.15
    ../../
    0.12
    ãģĦãģĨ
    0.10
    窿
    0.10
    entes
    0.09
    hots
    0.09
    undi
    0.09
    idad
    0.09
    ting
    0.09
    IDAD
    0.09
    Activations Density 0.015%

    No Known Activations