© 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. 58126
    Prev
    Next
    INDEX
    Explanations

    public domain

    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
    åѤç«ĭ
    -0.10
    вал
    -0.09
    .asp
    -0.09
    obl
    -0.09
    åłµ
    -0.08
    olest
    -0.08
    çĮ¿
    -0.08
    CAF
    -0.08
    çģĮ
    -0.07
     Islands
    -0.07
    POSITIVE LOGITS
    ãĥķãĥªãĥ¼
    0.12
     Ñģвобод
    0.12
    èĩªçĶ±
    0.12
     free
    0.12
     libre
    0.11
    -free
    0.11
    ä¸įåıĹ
    0.11
    /free
    0.11
    fre
    0.10
    free
    0.10
    Activations Density 0.153%

    No Known Activations