© 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. 2276
    Prev
    Next
    INDEX
    Explanations

    Niagara and Nicaragua

    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
    æľ¬ç½ij
    -0.10
    åĽĬ
    -0.09
    itch
    -0.09
    iah
    -0.09
    rics
    -0.08
    nis
    -0.08
    asing
    -0.08
    iliary
    -0.08
    æ¬
    -0.08
    ister
    -0.08
    POSITIVE LOGITS
    çĢijå¸ĥ
    0.11
     Heights
    0.10
    YPE
    0.10
    à¹Īาà¸ĩ
    0.09
    éĿŀåĩ¡
    0.09
    urses
    0.09
    ä¹ħ
    0.08
    lage
    0.08
    两岸
    0.08
    dale
    0.08
    Activations Density 0.002%

    No Known Activations