© 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. 31804
    Prev
    Next
    INDEX
    Explanations

    constituent or constituency

    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
    -0.09
    sam
    -0.09
    åĴĮå®ĮåĸĦ
    -0.09
    ToInt
    -0.08
    ä¿£
    -0.08
    éŃĦ
    -0.08
    å®ķ
    -0.08
    icult
    -0.08
     initials
    -0.08
    使ãģĦ
    -0.08
    POSITIVE LOGITS
    rab
    0.11
    åºĵéĩĮ
    0.10
    å°±ä¸ļ
    0.09
    поÑĤ
    0.09
    ello
    0.09
    åºĵ
    0.09
    èĥ¡èIJĿåįľ
    0.09
    fra
    0.09
    亲åıĭ
    0.09
    åŁĥ
    0.09
    Activations Density 0.148%

    No Known Activations