© 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. 17443
    Prev
    Next
    INDEX
    Explanations

    than comparisons

    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
     att
    -0.10
    ork
    -0.09
     неÑģколÑĮкиÑħ
    -0.08
    ĽĦ
    -0.08
    åĿ¦
    -0.08
    à¹ĥ
    -0.08
    äºĮæĺ¯
    -0.08
    hof
    -0.08
    hydrate
    -0.08
    ниÑĩ
    -0.08
    POSITIVE LOGITS
    æľ¬ç½ij
    0.10
    .wrapper
    0.09
    åºĶæľīçļĦ
    0.09
    cents
    0.09
    èĤĩ
    0.09
    ä»»ä½ķ人
    0.08
    æľīä»»ä½ķ
    0.08
    çļĦå±ĢéĿ¢
    0.08
    å½ĵæŶ
    0.08
    akit
    0.08
    Activations Density 0.068%

    No Known Activations