© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 49664
    Prev
    Next
    INDEX
    Explanations

    instruction pipeline processor

    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
    å¼ĢæĪ·
    -0.10
    彩
    -0.09
    ä¸Ī
    -0.09
    _MARGIN
    -0.09
    éģĩ
    -0.09
    è§ģè¯ģ
    -0.09
    累累
    -0.09
    oled
    -0.08
    çļĦç»ıåİĨ
    -0.08
    ransition
    -0.08
    POSITIVE LOGITS
     instruction
    0.09
    被æī§è¡Į
    0.09
     Intr
    0.08
    intr
    0.08
     intr
    0.08
    syscall
    0.08
     processor
    0.08
    /Instruction
    0.08
    struction
    0.08
    alary
    0.08
    Activations Density 0.322%

    No Known Activations