© 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. 19906
    Prev
    Next
    INDEX
    Explanations

    Lego buildings and blocks

    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
     каÑĩ
    -0.09
    è½®æµģ
    -0.09
    èĹ»
    -0.09
     parchment
    -0.09
     egreg
    -0.09
     каж
    -0.09
    ç´Ļ
    -0.09
     Coff
    -0.09
    两çĤ¹
    -0.09
    ê°ģ
    -0.08
    POSITIVE LOGITS
    ä¸įä¸Ĭ
    0.10
    è¡ĮæĶ¿æī§æ³ķ
    0.09
    åĿĹ
    0.09
    èĪĨ
    0.08
    sert
    0.08
    unset
    0.08
    inet
    0.08
    é¢Ĩ
    0.08
     blocks
    0.08
    blocks
    0.08
    Activations Density 0.028%

    No Known Activations