© 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. 58531
    Prev
    Next
    INDEX
    Explanations

    rose followed by mary

    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
    ä»Ĩ
    -0.11
    CTOR
    -0.10
    ities
    -0.09
    ance
    -0.09
    ÑĢаÑģ
    -0.09
    ä½ľé£İ
    -0.09
    è¬Ģ
    -0.09
    icc
    -0.09
    лиÑĨ
    -0.09
    ÑĢод
    -0.09
    POSITIVE LOGITS
    mary
    0.18
    MARY
    0.11
    idon
    0.11
    berry
    0.10
    dale
    0.10
    hips
    0.10
    boro
    0.10
    leaf
    0.09
    glyph
    0.09
    craft
    0.09
    Activations Density 0.032%

    No Known Activations