© 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. 32695
    Prev
    Next
    INDEX
    Explanations

    Copyright holder

    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
    å´ĩé«ĺ
    -0.09
    è¯ħ
    -0.09
    çIJIJ
    -0.09
     Recap
    -0.09
    å§ĭ
    -0.08
    :start
    -0.08
    ç¡Į
    -0.08
     ÑģлÑĥÑĩа
    -0.08
    alted
    -0.08
    ãģ¾ãĤĬ
    -0.08
    POSITIVE LOGITS
    åĭĥåĭĥ
    0.10
    inton
    0.10
    ç®Ń
    0.09
    Validity
    0.09
    å°Ħæīĭ
    0.09
    راء
    0.09
    /of
    0.08
    å¢ĥ
    0.08
    Via
    0.08
    oub
    0.08
    Activations Density 0.002%

    No Known Activations