© 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. 52878
    Prev
    Next
    INDEX
    Explanations

    json/code 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
    ваÑĢ
    -0.10
    tx
    -0.08
    ts
    -0.08
    avr
    -0.08
     Ñģобой
    -0.08
    uir
    -0.08
    more
    -0.08
    al
    -0.08
    åı¦ä¸ĢåįĬ
    -0.07
    tti
    -0.07
    POSITIVE LOGITS
     Mandal
    0.11
    lichkeit
    0.10
    _PCM
    0.09
    osten
    0.09
    å®ī稳
    0.09
    æ¤Ł
    0.09
    semblies
    0.09
    heels
    0.09
    ../../../
    0.09
    ests
    0.09
    Activations Density 0.318%

    No Known Activations