© 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. 9143
    Prev
    Next
    INDEX
    Explanations

    token suffixation

    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
    ures
    -0.11
    .LayoutParams
    -0.10
    ounding
    -0.10
    ðŁı»
    -0.09
    gun
    -0.09
    管å±Ģ
    -0.09
    maal
    -0.09
    roken
    -0.09
    unal
    -0.08
    alin
    -0.08
    POSITIVE LOGITS
    ized
    0.13
    omics
    0.10
    ization
    0.10
    RED
    0.09
    令çīĮ
    0.09
    isation
    0.09
    errupt
    0.09
    hm
    0.08
    iation
    0.08
    hs
    0.08
    Activations Density 0.037%

    No Known Activations