© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 7002
    Prev
    Next
    INDEX
    Explanations

    consent/refusal/domination dynamics

    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
     pare
    -0.08
    注åĨĮ
    -0.08
     Reco
    -0.08
    PWD
    -0.08
    æļĤç¼ĵ
    -0.08
     Pare
    -0.08
    .setImage
    -0.08
    åĽĬ
    -0.07
     plag
    -0.07
    pow
    -0.07
    POSITIVE LOGITS
    åıįæĬĹ
    0.14
    ä¹ĸä¹ĸ
    0.14
    被迫
    0.14
    被æĬĵ
    0.13
    被åĬ¨
    0.13
    éĢĥè·ij
    0.11
    驯
    0.11
    被æīĵ
    0.11
    ActionTypes
    0.10
    èĭ¦éļ¾
    0.10
    Activations Density 0.049%

    No Known Activations