© 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. 63479
    Prev
    Next
    INDEX
    Explanations

    ability and its forms

    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
     avis
    -0.10
    亮丽
    -0.08
    PathComponent
    -0.08
    ifle
    -0.08
    äºĮ代
    -0.08
    ivals
    -0.08
    upe
    -0.08
    Ctl
    -0.08
    rail
    -0.08
    ervice
    -0.08
    POSITIVE LOGITS
    -bodied
    0.11
    ÑģÑĤвоваÑĤÑĮ
    0.11
    æľīä¿¡å¿ĥ
    0.10
    ñana
    0.09
    autiful
    0.09
    azer
    0.09
     болÑĮÑĪое
    0.09
    æ°´å¹³
    0.09
    ноÑģÑĤ
    0.08
    èĪľ
    0.08
    Activations Density 0.026%

    No Known Activations