© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 10215
    Prev
    Next
    INDEX
    Explanations

    Felipe, Felicity, Felix

    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
    ingly
    -0.10
    matic
    -0.10
    ickle
    -0.10
    lined
    -0.09
    evil
    -0.09
    ä»¥äººä¸ºæľ¬
    -0.09
    lining
    -0.09
    enance
    -0.09
    sockets
    -0.09
    rade
    -0.09
    POSITIVE LOGITS
    istic
    0.11
    ials
    0.10
    ities
    0.10
    itim
    0.10
    	TokenName
    0.10
    ãĤĪãģĨãģª
    0.10
    ement
    0.09
    upy
    0.09
    еÑģÑĮ
    0.09
    ResourceManager
    0.09
    Activations Density 0.022%

    No Known Activations