© 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. 4022
    Prev
    Next
    INDEX
    Explanations

    light or lightweight descriptor

    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
    .jav
    -0.09
    hl
    -0.09
    antine
    -0.09
    å¶Ļ
    -0.09
    çIJ¨
    -0.09
    apel
    -0.08
    ois
    -0.08
    ä¸įä¸Ĭ
    -0.08
    rh
    -0.08
    PELL
    -0.08
    POSITIVE LOGITS
    /light
    0.15
    (light
    0.14
    weight
    0.14
    ning
    0.14
    -weight
    0.14
    weights
    0.14
    è½»
    0.13
    NING
    0.12
    nings
    0.12
    ened
    0.11
    Activations Density 0.195%

    No Known Activations