© 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. 18410
    Prev
    Next
    INDEX
    Explanations

    cadet, cadaver, cadmium, cadre, Cadigan

    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
    lix
    -0.10
    ariat
    -0.09
    abor
    -0.09
    ilege
    -0.09
    arring
    -0.08
    ä¸įå¼Ģ
    -0.08
    teenth
    -0.08
    亵
    -0.08
    æĹħ客
    -0.08
    è¢Ĥ
    -0.08
    POSITIVE LOGITS
    mium
    0.13
    и
    0.10
    iliar
    0.10
    cing
    0.09
    erno
    0.09
    enas
    0.09
    ural
    0.09
    ashi
    0.09
    tribute
    0.08
    rück
    0.08
    Activations Density 0.027%

    No Known Activations