© 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. 39734
    Prev
    Next
    INDEX
    Explanations

    bash commands and scripts

    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
    ally
    -0.10
    ç»İ
    -0.09
    ÑĪе
    -0.08
    uct
    -0.08
    ju
    -0.08
    ately
    -0.08
    SD
    -0.08
    ike
    -0.08
    åı£å²¸
    -0.08
    apolis
    -0.07
    POSITIVE LOGITS
    	TokenName
    0.13
    ëģĶ
    0.13
    busters
    0.11
    íĴĪ
    0.11
    ãĤĤãģ£ãģ¨
    0.10
    (es
    0.10
    away
    0.10
    nock
    0.10
    iaux
    0.10
    ableObject
    0.10
    Activations Density 0.008%

    No Known Activations