© 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. 38038
    Prev
    Next
    INDEX
    Explanations

    poorly framed concepts

    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
    çķĪ
    -0.09
     rare
    -0.09
    _BAR
    -0.09
     parcel
    -0.08
    é²ľæ´»
    -0.08
    éĵº
    -0.08
    undi
    -0.08
     Hayden
    -0.08
    并ä¸İ
    -0.08
     temperament
    -0.08
    POSITIVE LOGITS
    jack
    0.11
    è¯ī
    0.10
    åºĶä»ĺ
    0.10
    åĭī强
    0.10
    tit
    0.10
    пло
    0.09
    alc
    0.09
     (\<
    0.09
    å¼±
    0.09
    åıªä¸º
    0.09
    Activations Density 0.144%

    No Known Activations