© 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. 50985
    Prev
    Next
    INDEX
    Explanations

    sub menu or branches

    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
    èįī
    -0.09
    èĢķèĢĺ
    -0.09
    éĴī
    -0.09
     Grass
    -0.09
    汩
    -0.08
    èįĴ
    -0.08
     Rog
    -0.08
     Ting
    -0.08
    _soft
    -0.08
    ney
    -0.08
    POSITIVE LOGITS
     branches
    0.12
    æłijæŀĿ
    0.12
    branches
    0.11
    ÑĤÑĢо
    0.10
    åĵģçīĮçļĦ
    0.10
    -unstyled
    0.09
    ç»ĵæŀĦè°ĥæķ´
    0.09
    _branch
    0.09
    央
    0.09
    .branch
    0.09
    Activations Density 0.163%

    No Known Activations