© 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. 50004
    Prev
    Next
    INDEX
    Explanations

    incredible thing

    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
    rooms
    -0.11
    OrCreate
    -0.10
    manship
    -0.10
    room
    -0.10
    est
    -0.09
    este
    -0.09
    ilden
    -0.08
    enticator
    -0.08
    lon
    -0.08
    ings
    -0.08
    POSITIVE LOGITS
     Hulk
    0.12
    íŀĪ
    0.11
    ìĶ©
    0.11
    ment
    0.11
    窿
    0.10
    åĪ©äºļ
    0.10
    ãģĹãģ
    0.10
    egis
    0.10
    é³
    0.10
    esimal
    0.10
    Activations Density 0.004%

    No Known Activations