© 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. 36853
    Prev
    Next
    INDEX
    Explanations

    parts or units

    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
    ship
    -0.13
    ships
    -0.12
    shire
    -0.11
    çĹĩ
    -0.10
    åĹ£
    -0.09
    éĺ³åĮº
    -0.09
    rence
    -0.09
    à¹Į
    -0.09
    rec
    -0.09
    lander
    -0.09
    POSITIVE LOGITS
    åĦ¿
    0.15
    wise
    0.12
    åħĴ
    0.12
    -sama
    0.10
    ARENT
    0.10
    hips
    0.10
    antry
    0.09
    zelf
    0.09
    tures
    0.09
    urement
    0.09
    Activations Density 0.022%

    No Known Activations