© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 9597
    Prev
    Next
    INDEX
    Explanations

    Curtis followed by a name or descriptor

    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
    éļĻ
    -0.10
    رة
    -0.09
    ¤
    -0.09
    缮åħī
    -0.09
    lege
    -0.09
    以æѤ
    -0.08
    ingly
    -0.08
    襻
    -0.08
     Cheney
    -0.08
    ert
    -0.08
    POSITIVE LOGITS
    obl
    0.11
    IRS
    0.10
    cript
    0.09
    ANCE
    0.09
    amel
    0.09
    éĻIJãĤĬ
    0.09
    ãĥĵãĥ«
    0.09
    ÐľÐĺ
    0.08
    prung
    0.08
    ahi
    0.08
    Activations Density 0.006%

    No Known Activations