© 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. 4519
    Prev
    Next
    INDEX
    Explanations

    prefixes that lead to entity names

    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
    quier
    -0.10
    æĵį
    -0.10
    umat
    -0.09
    åħħåĪĨ
    -0.08
    CAST
    -0.08
    rom
    -0.08
     Leg
    -0.08
    åĨľåī¯äº§åĵģ
    -0.08
    iz
    -0.08
     suffice
    -0.08
    POSITIVE LOGITS
    åĴĮæĬĢæľ¯
    0.10
    æŶèĬĤ
    0.10
    äºĭåħĪ
    0.10
    ä¸ĩçī©
    0.09
    åİļåİļ
    0.09
    åĮĹæŀģ
    0.09
    åĻ
    0.09
    åįļè§Ī
    0.08
    è¿ijè·Ŀ离
    0.08
    å½ĵåľº
    0.08
    Activations Density 0.010%

    No Known Activations