© 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. 21485
    Prev
    Next
    INDEX
    Explanations

    referred to as

    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
    ATED
    -0.08
    tel
    -0.08
    erson
    -0.08
    rons
    -0.08
    ----------------------------------------------------------------------------
    -0.08
     rừng
    -0.08
     procrast
    -0.08
    ä¸Ģå®ļçļĦ
    -0.08
    ä¸į失
    -0.08
    iert
    -0.08
    POSITIVE LOGITS
    ä¸įä¸Ĭ
    0.10
    ase
    0.10
    _TC
    0.10
    ìĤ¬íķŃ
    0.09
    noinspection
    0.09
    encing
    0.09
    rence
    0.09
    ðŁı»
    0.09
    ake
    0.09
    pen
    0.08
    Activations Density 0.031%

    No Known Activations