© 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. 1000
    Prev
    Next
    INDEX
    Explanations

    Roberts

    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
    OrUpdate
    -0.11
    第äºĮå±Ĭ
    -0.10
    åĮĸè¿Ľç¨ĭ
    -0.09
    èĻ±
    -0.09
    èįĴ
    -0.09
    å¸ħ
    -0.09
    yard
    -0.09
    éĺĸ
    -0.09
    红æĹĹ
    -0.09
    æĪĺçķ¥å¸ĥå±Ģ
    -0.09
    POSITIVE LOGITS
    aceous
    0.10
    5
    0.09
     Brace
    0.09
    4
    0.08
    0
    0.08
    3
    0.08
    itage
    0.08
    7
    0.08
    avn
    0.08
    (EFFECT
    0.08
    Activations Density 0.003%

    No Known Activations