© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 20434
    Prev
    Next
    INDEX
    Explanations

    saddle or related term

    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
    aceous
    -0.10
    ince
    -0.10
    å¼łè´´
    -0.09
    aksi
    -0.09
    ally
    -0.09
    chem
    -0.08
    ening
    -0.08
     smb
    -0.08
    SA
    -0.08
    æŀľ
    -0.08
    POSITIVE LOGITS
    /or
    0.13
    /OR
    0.11
    æģ¬
    0.10
    座
    0.10
    /runtime
    0.09
    éŀ´
    0.09
    æ©Ľ
    0.09
    -shaped
    0.09
    åĪĽæĬķ
    0.09
    lace
    0.09
    Activations Density 0.004%

    No Known Activations