© 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. 56136
    Prev
    Next
    INDEX
    Explanations

    tackle problems

    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
    est
    -0.14
    nings
    -0.11
    ening
    -0.11
     rÃłng
    -0.10
    gie
    -0.09
    æĩĭ
    -0.09
    estre
    -0.09
    ply
    -0.09
    kowski
    -0.09
    lie
    -0.09
    POSITIVE LOGITS
    OutOfBounds
    0.09
    ä¿Ŀé²ľ
    0.09
    çĶŁèĤĸ
    0.09
    lements
    0.08
    bars
    0.08
    :[[
    0.08
    OutOfRangeException
    0.08
    olls
    0.08
    çķĶ
    0.08
    è¾į
    0.08
    Activations Density 0.003%

    No Known Activations