© 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. 19488
    Prev
    Next
    INDEX
    Explanations

    ends in "-ine", "-erie", "-ate", "-ily"

    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
    çħ
    -0.11
    çĬĩ
    -0.10
    æħ
    -0.09
    /arch
    -0.09
    çļĦé£Łçī©
    -0.09
    egas
    -0.09
    被æī§è¡Į
    -0.08
    orra
    -0.08
    æŃ£æĺ¯åĽłä¸º
    -0.08
    å±ħä½ı
    -0.08
    POSITIVE LOGITS
    anine
    0.11
    erie
    0.10
    irate
    0.09
    arily
    0.09
    åĸĥ
    0.09
     sund
    0.09
    tsy
    0.08
    asty
    0.08
    eries
    0.08
    éĺ¶çº§
    0.08
    Activations Density 0.143%

    No Known Activations