© 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. 35075
    Prev
    Next
    INDEX
    Explanations

    low followed by descriptive word

    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
    SSION
    -0.11
    maal
    -0.09
    MENT
    -0.09
    alog
    -0.09
    weit
    -0.08
    باط
    -0.08
    SCII
    -0.08
    è¶ħé«ĺ
    -0.08
    ady
    -0.08
    ÑĤел
    -0.08
    POSITIVE LOGITS
    ening
    0.14
    ish
    0.13
    à¹Ĩ
    0.13
    est
    0.13
    Enough
    0.11
     à¹Ĩ
    0.11
     enough
    0.11
    /simple
    0.11
    (er
    0.11
    ens
    0.10
    Activations Density 0.035%

    No Known Activations