© 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. 21115
    Prev
    Next
    INDEX
    Explanations

    double, dual, triple, -fold multipliers

    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
    SplitOptions
    -0.10
    RIES
    -0.09
    .present
    -0.09
    DIRECTORY
    -0.08
    .precision
    -0.08
    íĺij
    -0.08
     chÃŃ
    -0.08
    /Runtime
    -0.08
    poses
    -0.08
     Tits
    -0.08
    POSITIVE LOGITS
    åıĮéĩį
    0.25
    å¤ļéĩį
    0.14
     dual
    0.14
     double
    0.14
    åıĮ
    0.13
     triple
    0.12
    åħ¼åħ·
    0.12
     doubly
    0.12
    éĽĻ
    0.12
    两æīĭ
    0.11
    Activations Density 0.071%

    No Known Activations