© 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. 58351
    Prev
    Next
    INDEX
    Explanations

    engraving, etching, embossed

    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
     Ñģли
    -0.10
    rik
    -0.10
    GetMethod
    -0.09
    _DELTA
    -0.09
    hood
    -0.08
    news
    -0.08
     sticky
    -0.08
    tha
    -0.08
    uters
    -0.08
    vlc
    -0.08
    POSITIVE LOGITS
    çIJ¢
    0.12
    etch
    0.11
    åĪ»
    0.11
    潦
    0.10
    etched
    0.10
    ä¸īåįĥ
    0.10
    ALLE
    0.09
    å¾Īæ·±
    0.09
    jee
    0.09
    OrUpdate
    0.09
    Activations Density 0.016%

    No Known Activations