© 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. 18227
    Prev
    Next
    INDEX
    Explanations

    faulty, defective, flawed

    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
    çĥŃçĤ¹
    -0.10
    OTE
    -0.10
    çİĦ
    -0.09
    phans
    -0.09
    dings
    -0.09
    .opensource
    -0.09
    finger
    -0.08
    chest
    -0.08
    ragen
    -0.08
    æĦī
    -0.08
    POSITIVE LOGITS
    æĪĸèĢħ说
    0.13
    è¿ĩ硬
    0.09
     Jefferson
    0.09
    avad
    0.09
     flawed
    0.09
    клÑİÑĩен
    0.09
    ated
    0.09
    ities
    0.08
    èĽĭ
    0.08
    æį¢ä¸ª
    0.08
    Activations Density 0.036%

    No Known Activations