© 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. 10167
    Prev
    Next
    INDEX
    Explanations

    sharpness, spine, thorn, barb

    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
     Wan
    -0.10
    ä¾£
    -0.09
    åĤ»
    -0.09
    çĶŁæľº
    -0.09
    æĭį
    -0.09
    认
    -0.09
    è¿Ļä¹Īä¹ħ
    -0.09
     Plenty
    -0.08
    åħįç¨İ
    -0.08
    ibi
    -0.08
    POSITIVE LOGITS
    åĪº
    0.17
    éĴĪ
    0.15
     prick
    0.14
    éĩĿ
    0.13
    å°ĸ
    0.13
    æ£ĺ
    0.13
    éĴī
    0.12
    éĶIJ
    0.12
     shar
    0.12
    æīİ
    0.11
    Activations Density 0.066%

    No Known Activations