© 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. 26769
    Prev
    Next
    INDEX
    Explanations

    adjectives describing physical qualities

    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
    oret
    -0.10
    hip
    -0.09
    ÑĪка
    -0.09
    /rem
    -0.09
    好åĿı
    -0.09
    ÑĤен
    -0.08
    ÑĤел
    -0.08
    кÑĢаÑĤ
    -0.08
    irs
    -0.08
    reb
    -0.08
    POSITIVE LOGITS
    ness
    0.21
    ening
    0.20
    à¹Ĩ
    0.17
     enough
    0.17
    NESS
    0.16
     à¹Ĩ
    0.16
    ened
    0.16
    (er
    0.16
    est
    0.15
     جدا
    0.14
    Activations Density 0.112%

    No Known Activations