© 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. 5195
    Prev
    Next
    INDEX
    Explanations

    h followed by an ending

    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
    ities
    -0.12
    ehen
    -0.11
    ech
    -0.11
    hev
    -0.10
    že
    -0.10
    ow
    -0.10
    usch
    -0.10
    heading
    -0.10
    HEMA
    -0.10
    SSION
    -0.10
    POSITIVE LOGITS
    s
    0.15
    enticator
    0.13
    udson
    0.13
    enticate
    0.12
    entication
    0.12
    ierarchy
    0.11
    azard
    0.11
    essian
    0.11
    itecture
    0.11
    allenge
    0.11
    Activations Density 0.589%

    No Known Activations