© 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. 6878
    Prev
    Next
    INDEX
    Explanations

    works by

    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
    ois
    -0.10
    è´¨
    -0.09
    鸥
    -0.09
    æŀľæłij
    -0.09
    éĥ¨éĺŁ
    -0.09
    lav
    -0.09
    åĭĩæ°Ķ
    -0.08
    ложениÑı
    -0.08
    éľĪ
    -0.08
    town
    -0.08
    POSITIVE LOGITS
    rix
    0.10
    educ
    0.09
    itten
    0.09
    setDefault
    0.09
    statt
    0.09
    ertz
    0.09
    itecture
    0.08
    udu
    0.08
    odo
    0.08
    nut
    0.08
    Activations Density 0.037%

    No Known Activations