© 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. 10665
    Prev
    Next
    INDEX
    Explanations

    narcissistic or narcotic

    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
    erval
    -0.10
    iment
    -0.10
    jid
    -0.09
    ild
    -0.09
    宵
    -0.09
    noch
    -0.09
    为人
    -0.08
    forth
    -0.08
    æŃĩ
    -0.08
    ifferent
    -0.08
    POSITIVE LOGITS
    eer
    0.10
    istic
    0.10
    ism
    0.09
    èĥ¥
    0.09
    çĽ¯
    0.08
    æ¯Ļ
    0.08
    us
    0.08
    ITY
    0.08
    çĥģ
    0.08
    arro
    0.08
    Activations Density 0.004%

    No Known Activations