© 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. 33298
    Prev
    Next
    INDEX
    Explanations

    license and version information

    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
    -quarters
    -0.14
    rd
    -0.12
    ãģıãĤĮ
    -0.11
    «
    -0.10
    ement
    -0.10
    åĴĮåľ°åĮº
    -0.10
    tern
    -0.09
    å²Ķ
    -0.09
    ly
    -0.09
    ubo
    -0.09
    POSITIVE LOGITS
    ãģĹãģ¾ãģĨ
    0.20
    ãģĹãģ¾
    0.11
    ively
    0.11
    åīįæīĢ
    0.10
    ëį°
    0.10
    ãģ¤ãģĭ
    0.10
    ivity
    0.10
    ãģĵãģ¨ãģĮãģ§ãģįãĤĭ
    0.10
    ors
    0.10
    ilion
    0.09
    Activations Density 0.285%

    No Known Activations