© 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. 22253
    Prev
    Next
    INDEX
    Explanations

    Reeves, Reeve, creeper, elected

    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
    ovan
    -0.09
    ronic
    -0.09
    è¯·ä½ł
    -0.09
    ruk
    -0.09
    تÙĪر
    -0.09
    fac
    -0.09
    侯
    -0.08
    лаÑģ
    -0.08
    æ¡ķ
    -0.08
     cand
    -0.08
    POSITIVE LOGITS
    es
    0.10
    utive
    0.09
    SIDE
    0.09
    ãĥIJãĥĥãĤ¯
    0.08
    eger
    0.08
    社
    0.08
    è°¤
    0.08
     crud
    0.08
    ely
    0.08
    esser
    0.08
    Activations Density 0.005%

    No Known Activations