© 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. 31984
    Prev
    Next
    INDEX
    Explanations

    hypothesis

    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
    嵬
    -0.09
    æĵį
    -0.09
     اÙĦتÙĤ
    -0.08
     affairs
    -0.08
    гин
    -0.08
    ousel
    -0.08
    ORAGE
    -0.08
    oyal
    -0.08
    ibli
    -0.08
    jin
    -0.08
    POSITIVE LOGITS
     hypothesis
    0.12
     hypotheses
    0.12
    é¢ĦæľŁ
    0.10
     backed
    0.10
    mas
    0.10
     tested
    0.10
    é¢Ħè¨Ģ
    0.09
    tested
    0.09
    ç¿ģ
    0.09
    amber
    0.09
    Activations Density 0.109%

    No Known Activations