© 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. 27101
    Prev
    Next
    INDEX
    Explanations

    quoted names and 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
    ÑĢаг
    -0.08
     Encounter
    -0.08
     à¹Ĩ
    -0.08
    /current
    -0.08
    ush
    -0.08
    .lex
    -0.08
    halt
    -0.08
     tern
    -0.08
    à¹Ģ
    -0.07
    æ²Ķ
    -0.07
    POSITIVE LOGITS
    æĿ¥ä¸įåıĬ
    0.10
    åŁļ
    0.09
    å¹£
    0.09
    为人
    0.09
    appable
    0.08
    oven
    0.08
     Kraft
    0.08
    ombo
    0.08
    ippet
    0.08
     Tata
    0.08
    Activations Density 0.051%

    No Known Activations