© 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. 5705
    Prev
    Next
    INDEX
    Explanations

    and/rand followed by common 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
    ends
    -0.11
    ting
    -0.11
    ches
    -0.10
    isted
    -0.10
    ists
    -0.09
    社ä¼ļ稳å®ļ
    -0.09
    åıijå±ķ
    -0.09
     íĨµíķ´ìĦľ
    -0.09
    BOARD
    -0.09
    ionale
    -0.09
    POSITIVE LOGITS
    /or
    0.23
    /OR
    0.19
    rew
    0.13
    reas
    0.12
    rogen
    0.11
    rade
    0.11
    Ñģком
    0.10
    ean
    0.10
    rea
    0.10
    sworth
    0.10
    Activations Density 0.069%

    No Known Activations