© 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. 62726
    Prev
    Next
    INDEX
    Explanations

    similarly or likewise

    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
    éĤ£ä¸ĢåĪ»
    -0.09
    åĵĤ
    -0.09
    owed
    -0.09
    ows
    -0.09
    åħ³éĶ®æĺ¯
    -0.09
    åĪ¹
    -0.09
    ä¸ĢåĪĨéĴŁ
    -0.09
    ä»ĸå¦Ī
    -0.09
    缮çļĦæĺ¯
    -0.09
    èľĤ
    -0.09
    POSITIVE LOGITS
    rap
    0.10
    -sized
    0.10
    å£ķ
    0.09
    _CHUNK
    0.09
     Voll
    0.08
    eselect
    0.08
    è¾¾å°Ķ
    0.08
    hood
    0.08
    ily
    0.08
    æĿ²
    0.08
    Activations Density 0.011%

    No Known Activations