© 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. 32424
    Prev
    Next
    INDEX
    Explanations

    reset or resetting

    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
     Gram
    -0.09
    éĻĦåĬłåĢ¼
    -0.08
    éĽ¾
    -0.08
    ives
    -0.08
    lint
    -0.08
    첨
    -0.08
    shed
    -0.08
    _CY
    -0.08
    мел
    -0.07
    å¦Ĩ
    -0.07
    POSITIVE LOGITS
    tings
    0.13
    tle
    0.11
    ting
    0.11
    tlement
    0.11
    OrCreate
    0.10
    åĪ°æŶåĢĻ
    0.10
     sÃłng
    0.10
    itial
    0.09
    éĥ½è®¤ä¸º
    0.09
    ted
    0.09
    Activations Density 0.069%

    No Known Activations