© 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. 5481
    Prev
    Next
    INDEX
    Explanations

    protection, prepare, security, Trust

    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
    èĥľåĪ©
    -0.12
    пÑĢÑı
    -0.11
    æĶ
    -0.11
    åıĹ害èĢħ
    -0.10
    åħĭæľį
    -0.10
    pan
    -0.10
    _Instance
    -0.09
    _sw
    -0.09
    altar
    -0.09
    riere
    -0.09
    POSITIVE LOGITS
    稳妥
    0.09
    æģ¼
    0.09
    agon
    0.08
    ä¿Ŀ
    0.08
    飬
    0.08
    åį±éĻ©
    0.08
    纵横
    0.08
    å¹²çĩ¥
    0.08
     protective
    0.08
    Protect
    0.08
    Activations Density 0.051%

    No Known Activations