© 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. 56672
    Prev
    Next
    INDEX
    Explanations

    malicious attacks and content

    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
    urer
    -0.10
    aleb
    -0.09
    blade
    -0.09
    lify
    -0.09
    ä¹ĭåĬ¿
    -0.09
    éģµå®Ī
    -0.08
    çĪ·
    -0.08
    åĽłä¸ºå®ĥ
    -0.08
    _APPRO
    -0.08
    wares
    -0.08
    POSITIVE LOGITS
    ally
    0.10
    åħħ足
    0.08
    emin
    0.08
    寡
    0.08
    ï¸ı
    0.08
    æݨå¼Ģ
    0.08
    å¿ĸ
    0.08
    åīįèĢħ
    0.08
    à¤
    0.08
    aire
    0.08
    Activations Density 0.002%

    No Known Activations