© 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. 7425
    Prev
    Next
    INDEX
    Explanations

    training

    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¶æĺ¯
    -0.13
    ter
    -0.10
    æĭĽçīĮ
    -0.10
    akin
    -0.09
    uate
    -0.09
    است
    -0.09
    个å¤ļ
    -0.08
    atis
    -0.08
    èķ¤
    -0.08
    oks
    -0.08
    POSITIVE LOGITS
    ä¸İåıijå±ķ
    0.11
    ocate
    0.10
    åĴĮåıijå±ķ
    0.09
    imony
    0.09
    imonial
    0.09
    å½Ģ
    0.09
    happy
    0.08
    bps
    0.08
    anzi
    0.08
    -kit
    0.08
    Activations Density 0.036%

    No Known Activations