© 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. 19571
    Prev
    Next
    INDEX
    Explanations

    Sacco, Cooper, Giudice; names

    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
     заÑĢ
    -0.10
    æĪĸåħ¶ä»ĸ
    -0.09
    ä»ĭ
    -0.09
    å±ķå¼Ģ
    -0.08
    æijĺ
    -0.08
    -style
    -0.08
    åıĬåħ¶ä»ĸ
    -0.08
     оÑĢиг
    -0.08
    å¤ľæĻļ
    -0.08
    _APB
    -0.08
    POSITIVE LOGITS
    -and
    0.15
     &
    0.13
    ï¼Ĩ
    0.12
    _and
    0.11
    åĴĮ
    0.10
    两家
    0.10
    &m
    0.09
    à¹ģละ
    0.09
    èģĶ社
    0.09
    &B
    0.09
    Activations Density 0.471%

    No Known Activations