© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 21684
    Prev
    Next
    INDEX
    Explanations

    near, followed by specific words/phrases

    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
    ÄĻd
    -0.09
    оде
    -0.09
    -thirds
    -0.09
    /or
    -0.09
    ÙĢÙĢÙĢ
    -0.09
    èģĶç³»æĸ¹å¼ı
    -0.09
    çľ¼åħī
    -0.09
    eming
    -0.08
    ORTH
    -0.08
    åī©
    -0.08
    POSITIVE LOGITS
    à¹Ĩ
    0.12
     à¹Ĩ
    0.12
    shore
    0.11
    ness
    0.10
    abouts
    0.10
    åIJĦç§įåIJĦæł·
    0.09
    /at
    0.09
    /on
    0.09
    åIJĦç§į
    0.09
    à¹Ģà¸Ħ
    0.09
    Activations Density 0.020%

    No Known Activations