© 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. 2819
    Prev
    Next
    INDEX
    Explanations

    conflict between

    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
    æĶ¥
    -0.10
    quet
    -0.09
     Fallon
    -0.09
    quee
    -0.09
    å½ĵä¹ĭ
    -0.09
    åģıå·®
    -0.08
    ç¾ħ
    -0.08
    ials
    -0.08
    ä»ij
    -0.08
    rien
    -0.08
    POSITIVE LOGITS
    ingly
    0.11
    ãĤ¥
    0.10
    buster
    0.10
    ron
    0.08
    /generated
    0.08
    HashCode
    0.08
    /fixtures
    0.08
    too
    0.08
     оÑĤз
    0.08
    éĸ±
    0.08
    Activations Density 0.022%

    No Known Activations