© 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. 33612
    Prev
    Next
    INDEX
    Explanations

    titanium dioxide

    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
    ings
    -0.11
    vez
    -0.09
    ional
    -0.09
    ever
    -0.09
    ability
    -0.08
    zione
    -0.08
    vero
    -0.08
    ãĥ©ãĥ¼
    -0.08
    _formats
    -0.08
    ler
    -0.08
    POSITIVE LOGITS
     dioxide
    0.10
    idf
    0.09
    alley
    0.09
    etch
    0.09
    suit
    0.09
    ãģĨãģ¡ãģ«
    0.08
     Eli
    0.08
    argin
    0.08
    çĭĦ
    0.08
    storms
    0.08
    Activations Density 0.057%

    No Known Activations