© 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. 61207
    Prev
    Next
    INDEX
    Explanations

    physical impact

    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
    ties
    -0.09
    resses
    -0.08
     Geographic
    -0.08
     Juice
    -0.08
    ẻ
    -0.08
    伶
    -0.08
    Ẻ
    -0.08
    èĪIJ
    -0.08
    gro
    -0.08
    äºļ马
    -0.08
    POSITIVE LOGITS
    åĨ²åĩ»
    0.11
     impact
    0.11
    æĴŀåĩ»
    0.11
    impact
    0.10
    浪
    0.10
    çĭł
    0.09
    é¢Ĺ
    0.09
    åı«æĪij
    0.09
    jure
    0.09
    åªĴä»ĭ
    0.09
    Activations Density 0.063%

    No Known Activations