© 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. 30155
    Prev
    Next
    INDEX
    Explanations

    doing this

    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
    åıĪ称
    -0.08
    ç§Ĵ
    -0.08
    9
    -0.08
    è¿Ļ座
    -0.08
    æĢĢçĸij
    -0.08
    7
    -0.08
    0
    -0.08
    8
    -0.08
    åħįè´£
    -0.08
    4
    -0.07
    POSITIVE LOGITS
    ä¸ĩéĩĮ
    0.09
    ajo
    0.09
    <pre
    0.08
    _legal
    0.08
    çĽĨ
    0.08
    ego
    0.08
    ary
    0.08
    assi
    0.08
    eg
    0.08
    åį«
    0.08
    Activations Density 1.049%

    No Known Activations