© 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. 1607
    Prev
    Next
    INDEX
    Explanations

    percentages

    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
     оÑĤлиÑĩа
    -0.09
    èĩ³æѤ
    -0.09
    è¾ĸ
    -0.09
    AGER
    -0.09
    íĹĪ
    -0.09
    pta
    -0.08
    uy
    -0.08
    ä¸Ģ群
    -0.08
    åħ¶ä½Ļ
    -0.08
    èĩªæѤ
    -0.08
    POSITIVE LOGITS
    çĻ¾åĪĨä¹ĭ
    0.14
    ä¸īåĪĨä¹ĭä¸Ģ
    0.12
    ä¸Ģéĥ¨åĪĨ
    0.11
    çİĩè¾¾åĪ°
    0.11
    çİĩ
    0.10
    ä¸īåĪĨ
    0.10
    çİĩ为
    0.10
     dobr
    0.10
    çĶļèĩ³æĺ¯
    0.09
    çĻ¾åĪĨçĻ¾
    0.09
    Activations Density 0.156%

    No Known Activations