© 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. 19760
    Prev
    Next
    INDEX
    Explanations

    grounds, reasons, or charges

    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
    é¸
    -0.09
    زاÙĦ
    -0.08
    éĢ
    -0.08
    idl
    -0.08
    odel
    -0.08
    æīĢå¾Ĺ
    -0.08
    TEM
    -0.08
    樽
    -0.08
     presence
    -0.08
     Blanch
    -0.08
    POSITIVE LOGITS
     reasons
    0.21
    çļĦçIJĨçĶ±
    0.19
    çļĦåİŁåĽł
    0.18
     Reasons
    0.18
    çIJĨçĶ±
    0.17
    åİŁåĽł
    0.17
    çļĦç¼ĺæķħ
    0.16
    ä»Ģä¹ĪåİŁåĽł
    0.16
     ìĿ´ìľł
    0.15
    reason
    0.15
    Activations Density 0.051%

    No Known Activations