© 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. 32021
    Prev
    Next
    INDEX
    Explanations

    suite number

    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
    oven
    -0.10
    ä¸Ģåij³
    -0.10
     XCT
    -0.09
    eres
    -0.09
    crate
    -0.08
    леп
    -0.08
    emax
    -0.08
    èī°
    -0.08
    åľ¨å®¶éĩĮ
    -0.08
    ç½ij讯
    -0.08
    POSITIVE LOGITS
    -suite
    0.10
    suite
    0.09
    Suite
    0.09
    ä¸Ģå±Ĥ
    0.08
     Suite
    0.08
    itch
    0.08
     Anc
    0.08
    楹
    0.08
    ub
    0.08
    CommandEvent
    0.08
    Activations Density 0.023%

    No Known Activations