© 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. 10523
    Prev
    Next
    INDEX
    Explanations

    index followed by file extensions or delimiters

    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
    CodeAt
    -0.11
    elden
    -0.10
    illage
    -0.10
    ties
    -0.09
    è¯·ä½ł
    -0.09
    å¼Ĥåľ°
    -0.08
    PED
    -0.08
    irty
    -0.08
    ips
    -0.08
    cheduler
    -0.08
    POSITIVE LOGITS
    OutOfBoundsException
    0.11
    lam
    0.09
    hibit
    0.09
    人士
    0.09
    edList
    0.09
    adores
    0.09
    ings
    0.09
    olated
    0.08
    ires
    0.08
    æĭ¾
    0.08
    Activations Density 0.063%

    No Known Activations