© 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. 18729
    Prev
    Next
    INDEX
    Explanations

    Pen ultimate

    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
    tte
    -0.09
    sWith
    -0.09
    sans
    -0.09
    ãĤªãĤ¹
    -0.08
    odem
    -0.08
    秣
    -0.08
    soc
    -0.08
    íij¸
    -0.08
     vegas
    -0.08
    snap
    -0.08
    POSITIVE LOGITS
    ning
    0.14
    ultimate
    0.14
    elope
    0.14
    etration
    0.11
    eral
    0.11
    ä¸ŃåĽ½å¸Ĥåľº
    0.10
    insula
    0.10
    åķ¶
    0.09
    elu
    0.09
    ormal
    0.09
    Activations Density 0.030%

    No Known Activations