© 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. 1041
    Prev
    Next
    INDEX
    Explanations

    executable files and commands

    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
    èµ·èįī
    -0.10
    tracked
    -0.08
    缮çļĦ
    -0.08
    deck
    -0.08
     Lub
    -0.08
    mnt
    -0.07
    缴æĴŃéĹ´
    -0.07
     xmlhttp
    -0.07
     Terror
    -0.07
     Hann
    -0.07
    POSITIVE LOGITS
    ipop
    0.10
    åĪ©åĻ¨
    0.10
    çĵł
    0.10
    íĴĪ
    0.10
     binaries
    0.10
    åĬŀäºĭ
    0.10
    å·¥åħ·
    0.10
    ç¥ŀåĻ¨
    0.09
    åĻ¨æ¢°
    0.09
    utable
    0.09
    Activations Density 0.246%

    No Known Activations