© 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. 51169
    Prev
    Next
    INDEX
    Explanations

    questions and requests

    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
    erb
    -0.10
    æł¹æľ¬å°±ä¸į
    -0.09
    èħĭ
    -0.09
    acks
    -0.09
    witter
    -0.08
    éģĦ
    -0.08
    è¹Ĥ
    -0.08
     заÑĢегиÑģÑĤ
    -0.08
    adian
    -0.08
    æľĭ
    -0.08
    POSITIVE LOGITS
    éĹ®æĪij
    0.10
    èĭŁ
    0.09
    rc
    0.09
     SPE
    0.09
     pose
    0.08
    éĹ®
    0.08
    åİī
    0.08
    éķ¿æķĪ
    0.08
    _SPE
    0.08
    spe
    0.08
    Activations Density 0.033%

    No Known Activations