© 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. 58093
    Prev
    Next
    INDEX
    Explanations

    active carbon fiber

    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
    füh
    -0.10
    SHOT
    -0.09
    éĮ
    -0.08
    Localized
    -0.08
    localized
    -0.08
    onde
    -0.08
    WithType
    -0.08
    æijĦ
    -0.08
    禹
    -0.08
    趺
    -0.08
    POSITIVE LOGITS
    åĬ²
    0.11
    èĮ§
    0.09
    åłij
    0.09
    第ä¸ĢæŃ¥
    0.09
    åĭģ
    0.08
    å¿ĥ缮
    0.08
    åĢ¼å¾Ĺä¸Ģ
    0.08
    èĭ¯
    0.08
    åŃĶ
    0.08
    çģ¼
    0.08
    Activations Density 0.423%

    No Known Activations