© 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. 6626
    Prev
    Next
    INDEX
    Explanations

    stream/streams

    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
    楯
    -0.10
    strap
    -0.10
    rees
    -0.09
     lẽ
    -0.09
    etch
    -0.09
    iable
    -0.08
    ÑģÑĤÑĢо
    -0.08
    	TokenName
    -0.08
    ORIA
    -0.08
    StreamReader
    -0.08
    POSITIVE LOGITS
    lining
    0.15
    ingly
    0.14
    ming
    0.12
    lined
    0.12
    side
    0.11
    lights
    0.10
    ÑĢ
    0.10
    atics
    0.10
    artin
    0.10
    rä
    0.10
    Activations Density 0.037%

    No Known Activations