© 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. 52482
    Prev
    Next
    INDEX
    Explanations

    evaluating, reproducing, adapting, generics

    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
    ed
    -0.25
    fully
    -0.13
    edu
    -0.12
    elijke
    -0.11
    edBy
    -0.11
    edo
    -0.11
    ted
    -0.11
    eners
    -0.11
    ish
    -0.10
    ened
    -0.10
    POSITIVE LOGITS
    SSION
    0.11
    usted
    0.10
    ainen
    0.10
    ção
    0.10
    ator
    0.10
    ations
    0.09
    ators
    0.09
    ATION
    0.09
    åŁ´
    0.09
    /mobile
    0.09
    Activations Density 0.278%

    No Known Activations