© 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. 39391
    Prev
    Next
    INDEX
    Explanations

    mono-

    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
    oting
    -0.09
    iner
    -0.09
    åıĸè¯ģ
    -0.09
    ough
    -0.09
    erman
    -0.09
    ovel
    -0.08
    ment
    -0.08
    ouse
    -0.08
    æĹłæķ°æ¬¡
    -0.08
    anke
    -0.08
    POSITIVE LOGITS
    æĪĸå¤ļ
    0.17
    /single
    0.16
    (single
    0.14
     baÅŁÄ±na
    0.12
    -single
    0.11
    tons
    0.11
    cip
    0.10
    ä¸ĸçķĮ级
    0.10
    çĵ¢
    0.10
    xes
    0.09
    Activations Density 0.121%

    No Known Activations