© 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. 3537
    Prev
    Next
    INDEX
    Explanations

    cautionary or precautionary

    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
    ãģĵãģ¨
    -0.11
    ski
    -0.10
    ularity
    -0.10
    å¢ĥçķĮ
    -0.09
    å¼ĥ
    -0.09
    æĥ³åĪ°
    -0.09
    shire
    -0.09
    大æĪ·
    -0.08
    lington
    -0.08
    pagen
    -0.08
    POSITIVE LOGITS
    iful
    0.10
    ately
    0.10
    磾
    0.10
    人éĢī
    0.10
    稳妥
    0.09
    ate
    0.09
    ably
    0.09
    tÃł
    0.09
    áb
    0.09
    å·²ä¹ħ
    0.09
    Activations Density 0.005%

    No Known Activations