© 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. 2323
    Prev
    Next
    INDEX
    Explanations

    bizarre followed by adjectives or nouns

    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
    ing
    -0.09
     пÑĥнк
    -0.09
    iore
    -0.09
    ingu
    -0.08
    atura
    -0.08
    eding
    -0.08
    race
    -0.08
    ungan
    -0.08
    tgt
    -0.08
    оваÑĢ
    -0.08
    POSITIVE LOGITS
    ly
    0.15
    LY
    0.12
    ously
    0.12
    оÑģÑĮ
    0.11
    ities
    0.11
    /random
    0.10
    eners
    0.10
    ä¸įçĦ¶
    0.10
    éĻ¬
    0.10
    媳
    0.10
    Activations Density 0.004%

    No Known Activations