© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 8365
    Prev
    Next
    INDEX
    Explanations

    hastily, eager, Haskell, HathiTrust, Hays Creek, execution

    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
    ault
    -0.11
    å³Ļ
    -0.10
    çªĴ
    -0.09
    ÑĢÑıд
    -0.09
    orest
    -0.09
    teenth
    -0.09
    l
    -0.09
    åľ°å¸¦
    -0.09
    çĬ¶æĢģä¸ĭ
    -0.09
    èIJ±
    -0.08
    POSITIVE LOGITS
    oppel
    0.10
    ое
    0.10
    ughs
    0.09
    forth
    0.09
    ively
    0.09
    robe
    0.09
    ayers
    0.09
    born
    0.09
    adium
    0.08
    ÃŃm
    0.08
    Activations Density 0.003%

    No Known Activations