© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 7480
    Prev
    Next
    INDEX
    Explanations

    appropriate

    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
    ings
    -0.12
    éĤij
    -0.11
    INGS
    -0.10
    ä¸ĩäºĭ
    -0.10
    mitt
    -0.09
    paces
    -0.09
    hack
    -0.09
    èĻ±
    -0.09
    lob
    -0.09
    -0.09
    POSITIVE LOGITS
    -sized
    0.13
    ities
    0.11
    å°º
    0.11
     sized
    0.11
     Sized
    0.11
    -looking
    0.11
    oop
    0.11
    ment
    0.10
    ately
    0.10
    âĶĬ
    0.09
    Activations Density 0.029%

    No Known Activations