© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 20860
    Prev
    Next
    INDEX
    Explanations

    fake news

    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
    åı¨
    -0.10
    éļħ
    -0.10
    igon
    -0.09
    à¹Į
    -0.09
    aries
    -0.09
    edImage
    -0.09
    Ñģол
    -0.09
     ÙĪÙĥÙĦ
    -0.09
    osal
    -0.09
    ILLE
    -0.09
    POSITIVE LOGITS
    able
    0.10
    ç»IJ
    0.10
    ULER
    0.10
    sters
    0.09
    abbrev
    0.09
    overs
    0.09
    -looking
    0.08
    ãĥ¼ãĤ·ãĥ§ãĥ³
    0.08
    nowledge
    0.08
    å¯IJ
    0.08
    Activations Density 0.014%

    No Known Activations