© 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. 30889
    Prev
    Next
    INDEX
    Explanations

    end followed by specific suffixes

    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
    °
    -0.11
    åıŁ
    -0.10
    anguard
    -0.10
    岸边
    -0.09
    ãĥ¼ãĥĪ
    -0.09
    ince
    -0.09
    asso
    -0.09
    etty
    -0.09
    ä¸ŃæĸŃ
    -0.09
     Serge
    -0.09
    POSITIVE LOGITS
    owment
    0.21
    urance
    0.18
    orses
    0.15
    angered
    0.15
    ding
    0.14
    ocrin
    0.14
    roit
    0.13
    ereco
    0.13
    ocrine
    0.12
    OfFile
    0.12
    Activations Density 0.099%

    No Known Activations