© 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. 60985
    Prev
    Next
    INDEX
    Explanations

    maps and layers data

    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
    æĮĤéĴ©
    -0.09
    _MPI
    -0.09
    ç¬Ļ
    -0.09
    imd
    -0.08
    çĤĬ
    -0.08
    indhoven
    -0.08
    缺
    -0.08
    以æĿ¥
    -0.08
    homes
    -0.08
    uja
    -0.08
    POSITIVE LOGITS
    å±Ģ
    0.09
    对åħ¬åı¸
    0.09
    å±¥
    0.09
     Shr
    0.09
     Trout
    0.09
    为人
    0.08
     Rut
    0.08
    hs
    0.08
    毫æĹłçĸij
    0.08
    .ReadAll
    0.08
    Activations Density 0.080%

    No Known Activations