© 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. 54480
    Prev
    Next
    INDEX
    Explanations

    state names, political offices, counties

    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
    èĬĴæŀľ
    -0.11
    æ¸Ŀ
    -0.11
    éªĸ
    -0.11
     Texas
    -0.10
    æķ¦çħĮ
    -0.10
    æĪIJéĥ½å¸Ĥ
    -0.10
    澳大åĪ©äºļ
    -0.10
    Texas
    -0.10
    éĢ¼
    -0.10
    客家
    -0.10
    POSITIVE LOGITS
     Connecticut
    0.18
     Massachusetts
    0.17
     Vermont
    0.16
     Boston
    0.15
     Maine
    0.14
    Boston
    0.14
    常å·ŀ
    0.13
     Celtics
    0.12
    åĵĪä½Ľ
    0.11
     Rhode
    0.11
    Activations Density 0.430%

    No Known Activations