© 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. 36540
    Prev
    Next
    INDEX
    Explanations

    convention and its follow-ups

    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
    sonian
    -0.10
    Ñģов
    -0.09
     constexpr
    -0.09
    éĶµ
    -0.09
    æľīå¾Ī大çļĦ
    -0.09
    饰
    -0.09
    owanie
    -0.08
    s
    -0.08
    -quarters
    -0.08
    姣
    -0.08
    POSITIVE LOGITS
    ally
    0.12
    ality
    0.11
    lasting
    0.11
    ary
    0.10
    aires
    0.10
    nel
    0.10
    chor
    0.09
    -era
    0.09
    ance
    0.09
    ANCE
    0.09
    Activations Density 0.022%

    No Known Activations