© 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. 32022
    Prev
    Next
    INDEX
    Explanations

    set-top box or TV

    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®Ľ
    -0.09
    cré
    -0.09
     Creek
    -0.09
    浸
    -0.08
    _ARCH
    -0.08
    ignet
    -0.08
    айн
    -0.08
    fruit
    -0.08
    ä¸Ŀ
    -0.08
    衬
    -0.08
    POSITIVE LOGITS
    çĶµè§Ĩ
    0.19
    ãĥĨãĥ¬ãĥĵ
    0.16
     television
    0.16
     Television
    0.16
    çľĭçĶµè§Ĩ
    0.15
     televis
    0.15
     TV
    0.14
    鼻è¦ĸ
    0.14
    çĶµè§Ĩæľº
    0.14
     tv
    0.13
    Activations Density 0.194%

    No Known Activations