© 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. 31650
    Prev
    Next
    INDEX
    Explanations

    im, implied, impress, impair

    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
    eros
    -0.12
    manship
    -0.12
    ings
    -0.11
     именно
    -0.11
    aways
    -0.10
    boom
    -0.10
    ìķ¼
    -0.09
    teen
    -0.09
    umb
    -0.09
    pard
    -0.09
    POSITIVE LOGITS
    agine
    0.15
    ãģ£ãģįãĤĬ
    0.12
    minent
    0.11
    æ·±åĪ»çļĦ
    0.11
    ments
    0.10
    balances
    0.10
    antly
    0.10
    ioned
    0.10
    mediately
    0.10
    oglob
    0.10
    Activations Density 0.097%

    No Known Activations