© 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. 10042
    Prev
    Next
    INDEX
    Explanations

    Warner Bros

    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
    åĥ§
    -0.10
    crest
    -0.09
    æģ©
    -0.09
    å¼¼
    -0.09
    éªı
    -0.08
    çĻ½éĽª
    -0.08
    ook
    -0.08
    çŃĿ
    -0.08
    æĶ¾è¿ĩ
    -0.08
    æĸ°ä¸ĸ纪
    -0.08
    POSITIVE LOGITS
    ä¿©
    0.11
    sWith
    0.10
    注æĦıåĬĽ
    0.09
    licher
    0.09
    ÑĤек
    0.09
    swith
    0.09
    ãĤ¯ãĥ©ãĥĸ
    0.08
    ï¸
    0.08
    ses
    0.08
    hawk
    0.08
    Activations Density 0.004%

    No Known Activations