© 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. 38908
    Prev
    Next
    INDEX
    Explanations

    frequently occurring adjectives

    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
    ed
    -0.10
     repost
    -0.09
    ary
    -0.09
    .ids
    -0.09
    èĤª
    -0.08
    ARY
    -0.08
    箪
    -0.08
    大è§Ħ模
    -0.08
    广æĴŃçĶµè§Ĩ
    -0.08
    æĬķ稿
    -0.08
    POSITIVE LOGITS
    istically
    0.18
    emente
    0.15
     обÑĢазом
    0.14
    ingly
    0.14
    aneously
    0.14
    fully
    0.13
    ently
    0.12
    entially
    0.12
    ively
    0.12
    lessly
    0.12
    Activations Density 0.291%

    No Known Activations