© 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. 48067
    Prev
    Next
    INDEX
    Explanations

    existing

    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
    mith
    -0.10
    éļħ
    -0.09
    å¯Ĥå¯ŀ
    -0.09
    aken
    -0.09
    ستاÙĨ
    -0.09
    sat
    -0.09
    akers
    -0.09
    æĪĸå¤ļ
    -0.09
    Asked
    -0.09
    å³Ļ
    -0.09
    POSITIVE LOGITS
    ence
    0.12
    entially
    0.11
    unger
    0.10
    fty
    0.10
    OrUpdate
    0.09
     ÙĦدÙĬÙĨا
    0.09
    ated
    0.09
    /generated
    0.09
    æĪĸå°Ĩ
    0.09
    ential
    0.09
    Activations Density 0.041%

    No Known Activations