© 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. 30181
    Prev
    Next
    INDEX
    Explanations

    bee followed by gees, keeping, street, or navigation

    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
    most
    -0.12
    ly
    -0.10
    sp
    -0.09
    ãģ£ãģįãĤĬ
    -0.09
    enko
    -0.09
    ation
    -0.09
    称ä¹ĭ
    -0.09
    edic
    -0.08
    ughter
    -0.08
    upt
    -0.08
    POSITIVE LOGITS
    hive
    0.13
    keepers
    0.10
    keeping
    0.09
    croft
    0.09
    饯
    0.09
    éĢĬ
    0.09
    eker
    0.09
    éķĿ
    0.09
    ONS
    0.09
    opard
    0.09
    Activations Density 0.008%

    No Known Activations