© 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. 22584
    Prev
    Next
    INDEX
    Explanations

    expand or expanding

    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
    uen
    -0.11
    åħµåĻ¨
    -0.10
    åĩºèº«
    -0.10
    å¾Īæ·±
    -0.09
    失æķĪ
    -0.09
    è¿ĩé«ĺ
    -0.09
    åĩºçĶŁ
    -0.09
    示èĮĥ
    -0.09
    æ¿ł
    -0.09
    æŁĦ
    -0.08
    POSITIVE LOGITS
    æī©å¼ł
    0.14
     expand
    0.14
    æī©å¤§
    0.14
    æī©
    0.13
     expands
    0.13
    -expand
    0.13
    æī©å±ķ
    0.13
    .expand
    0.13
     expansions
    0.12
     Expand
    0.12
    Activations Density 0.113%

    No Known Activations