© 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. 49709
    Prev
    Next
    INDEX
    Explanations

    x, ix, ax, bx

    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
    å©¿
    -0.12
    isci
    -0.11
     excess
    -0.11
    istic
    -0.11
    ared
    -0.10
    XXX
    -0.10
    ered
    -0.10
    XX
    -0.09
    ize
    -0.09
    X
    -0.09
    POSITIVE LOGITS
    plorer
    0.18
    tures
    0.16
    perience
    0.14
    avier
    0.14
    presso
    0.13
    ercise
    0.13
    cellent
    0.13
    ses
    0.12
    lsx
    0.12
    0.12
    Activations Density 0.607%

    No Known Activations