© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 9660
    Prev
    Next
    INDEX
    Explanations

    ratio

    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
     justified
    -0.09
     dope
    -0.09
    adian
    -0.09
    å¤©åľ°
    -0.08
     sorts
    -0.08
    为主
    -0.08
    ä¸ĸ家
    -0.08
    å°ı康
    -0.08
    ä¸ĭè°ĥ
    -0.07
    2
    -0.07
    POSITIVE LOGITS
     ratio
    0.11
    ucker
    0.11
    æĩ
    0.10
    å§Ĭ
    0.10
    çĪ¬
    0.10
    平衡
    0.10
     Ratio
    0.10
    ratio
    0.09
    ::__
    0.09
    Ratio
    0.09
    Activations Density 0.065%

    No Known Activations