© 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. 17407
    Prev
    Next
    INDEX
    Explanations

    reference

    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
    åĬ¢
    -0.09
    piler
    -0.09
    uar
    -0.09
    çĥĪ
    -0.09
    çĽijçĿ£ç®¡çIJĨ
    -0.08
     ÐĴоз
    -0.08
    æĺ¯ä»ĸ
    -0.08
    è¯ĥ
    -0.08
    åķ¼
    -0.08
    æĺ¯æĢİæł·
    -0.08
    POSITIVE LOGITS
    åıĤçħ§
    0.13
     refer
    0.12
    æ±ĤåĬ©
    0.11
    åıĤèĢĥ
    0.11
    对çħ§
    0.11
    Refer
    0.11
     reference
    0.11
    refer
    0.10
    åıįå¤į
    0.10
     repeatedly
    0.10
    Activations Density 0.056%

    No Known Activations