© 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. 27997
    Prev
    Next
    INDEX
    Explanations

    jaw

    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
    èĻŀ
    -0.11
    ally
    -0.10
    ------+------+
    -0.10
    ively
    -0.09
    微信åı·
    -0.09
    ationally
    -0.09
    ikki
    -0.09
    anvas
    -0.09
    _pb
    -0.09
    tsy
    -0.09
    POSITIVE LOGITS
    bone
    0.12
    less
    0.11
    bones
    0.11
    骨
    0.10
    rende
    0.10
     Jaw
    0.10
    åħ·ä½ĵåĨħ容
    0.10
    ohl
    0.10
    /or
    0.09
    breaker
    0.09
    Activations Density 0.018%

    No Known Activations