© 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. 38265
    Prev
    Next
    INDEX
    Explanations

    Reuters

    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
    éĬ®
    -0.09
    side
    -0.09
    èī³
    -0.09
     кÑĥлÑĮÑĤÑĥÑĢÑĭ
    -0.08
    victim
    -0.08
    itore
    -0.08
    éļ½
    -0.08
    sey
    -0.08
    ensions
    -0.08
    licit
    -0.08
    POSITIVE LOGITS
    éĥ¨èIJ½
    0.08
    eneg
    0.08
    ize
    0.08
    hunt
    0.08
    lla
    0.08
    iza
    0.08
    èĻIJ
    0.08
     ìļ°ë¦¬ê°Ģ
    0.08
    .getM
    0.08
    å®īå¨ľ
    0.08
    Activations Density 0.005%

    No Known Activations