© 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. 32329
    Prev
    Next
    INDEX
    Explanations

    heavy and heavily

    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
    è¾Ī
    -0.10
    sWith
    -0.10
    afety
    -0.10
    ведиÑĤе
    -0.09
    yth
    -0.09
    cip
    -0.09
    eel
    -0.08
    byn
    -0.08
    sig
    -0.08
    cie
    -0.08
    POSITIVE LOGITS
    ening
    0.15
    -handed
    0.15
    -duty
    0.13
    ened
    0.13
    -hearted
    0.12
     mẽ
    0.12
    -bodied
    0.10
    ç»Ħç»ĩé¢Ĩ导
    0.10
    æĹłæ¯Ķ
    0.10
     hitters
    0.10
    Activations Density 0.012%

    No Known Activations