© 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. 19858
    Prev
    Next
    INDEX
    Explanations

    looking forward to

    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
    ä½łçľĭ
    -0.12
    ละ
    -0.09
    ination
    -0.09
    bjerg
    -0.09
    æģIJæħĮ
    -0.08
    째
    -0.08
    ä¼İ
    -0.08
    æµ£
    -0.08
    åIJĥè¿ĩ
    -0.08
    容
    -0.08
    POSITIVE LOGITS
    åıijå±ķä¸Ń
    0.09
    ãģļãģ£ãģ¨
    0.09
    京
    0.09
    нее
    0.09
    иÑĪ
    0.08
    åĩĽ
    0.08
    Powered
    0.08
     impunity
    0.08
    raise
    0.08
    ills
    0.08
    Activations Density 0.002%

    No Known Activations