© 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. 49287
    Prev
    Next
    INDEX
    Explanations

    pure followed by word

    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
     تعاÙĦÙī
    -0.10
    uster
    -0.10
    aber
    -0.10
    ewise
    -0.10
    uation
    -0.09
    re
    -0.09
    ç©°
    -0.09
    å¿¡
    -0.09
    naire
    -0.09
    rab
    -0.09
    POSITIVE LOGITS
    bred
    0.16
    zza
    0.12
    té
    0.11
    ÑģÑĮ
    0.11
    -hearted
    0.10
    foy
    0.10
    -blood
    0.10
    eing
    0.10
    äºĽ
    0.10
    going
    0.10
    Activations Density 0.010%

    No Known Activations