© 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. 19926
    Prev
    Next
    INDEX
    Explanations

    up, out, down movement

    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
    åĮĸè¿Ľç¨ĭ
    -0.11
    forth
    -0.10
     à¹Ĩ
    -0.09
    à¹Ĩ
    -0.09
    ized
    -0.09
    UpDown
    -0.09
    ised
    -0.09
    èĭ´
    -0.09
    Ñħод
    -0.09
    级以ä¸Ĭ
    -0.09
    POSITIVE LOGITS
    wards
    0.09
    2
    0.08
    6
    0.08
    è¿Ļå°±
    0.08
    /down
    0.08
     Reviewed
    0.08
    agner
    0.08
    ulin
    0.08
    oli
    0.08
    ferences
    0.08
    Activations Density 0.253%

    No Known Activations