© 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. 30784
    Prev
    Next
    INDEX
    Explanations

    ex following words like Media, Palestine, traction, omis, vivo, cretion, oplanes

    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
    inals
    -0.11
    ician
    -0.11
    ongs
    -0.11
    oret
    -0.10
     constexpr
    -0.10
    XXXX
    -0.09
    ieved
    -0.09
    atan
    -0.09
    XXX
    -0.09
    ovan
    -0.09
    POSITIVE LOGITS
    clamation
    0.13
    ively
    0.12
    ordinary
    0.11
    cluded
    0.11
    anding
    0.11
    един
    0.10
    cerpt
    0.10
    changer
    0.10
    PLICIT
    0.10
    auc
    0.10
    Activations Density 0.117%

    No Known Activations