© 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. 26925
    Prev
    Next
    INDEX
    Explanations

    animals

    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
    spo
    -0.11
    åħĪéĶĭ
    -0.10
    ä¸Ŀè·¯
    -0.10
    eneg
    -0.10
    çĥ½
    -0.09
    åĮĸèĤ¥
    -0.09
    enen
    -0.09
    åĽłåľ°
    -0.09
    agi
    -0.09
    CAPE
    -0.09
    POSITIVE LOGITS
    åĬ¨çī©
    0.14
    æµģ浪
    0.13
     animal
    0.13
    çıį
    0.11
    animal
    0.10
    æĭ¯æķij
    0.10
     Animal
    0.10
     animals
    0.10
    æķijåĬ©
    0.10
    åĭķçī©
    0.09
    Activations Density 0.308%

    No Known Activations