© 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. 1230
    Prev
    Next
    INDEX
    Explanations

    submarine types

    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
    çľĭ书
    -0.10
    å¹²çĩ¥
    -0.10
    åĩĿ
    -0.10
    ervo
    -0.10
     Capture
    -0.09
    ë§
    -0.09
    duk
    -0.09
    ลา
    -0.09
    åŀĽ
    -0.09
    ymoon
    -0.09
    POSITIVE LOGITS
    ç¥Ł
    0.09
    æ®ĥ
    0.09
    åĽ½éĻħè´¸æĺĵ
    0.08
    ç¿ģ
    0.08
     Aç
    0.08
    æİĴæ°Ķ
    0.08
     seperate
    0.08
     Schro
    0.08
    /mit
    0.08
     Licht
    0.08
    Activations Density 0.154%

    No Known Activations