© 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. 49011
    Prev
    Next
    INDEX
    Explanations

    regular CCP, sorted soil, large umbrella, main transition, supporting roll

    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
    çĤ¹è¯Ħ
    -0.09
    å®Įæ¯ķ
    -0.08
    _UTF
    -0.08
    å½ĵåľ°æĶ¿åºľ
    -0.08
    Finished
    -0.08
    UME
    -0.07
     Underground
    -0.07
    æłijç«ĭ
    -0.07
    unic
    -0.07
    å£ķ
    -0.07
    POSITIVE LOGITS
    家ä¼Ļ
    0.12
    /full
    0.09
    ness
    0.09
     же
    0.09
    /original
    0.09
    ussed
    0.09
    å±ĤéĿ¢
    0.09
    å¤ļåįĬ
    0.09
    bies
    0.09
    inia
    0.09
    Activations Density 0.627%

    No Known Activations