© 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. 62398
    Prev
    Next
    INDEX
    Explanations

    desire for features or outcomes

    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
    uration
    -0.10
    ìį¨
    -0.10
    rade
    -0.10
    rust
    -0.10
    Donald
    -0.09
    åĨĻåŃĹ
    -0.09
    river
    -0.09
    æıŃ
    -0.09
    rodu
    -0.08
    æĥ³æĥ³
    -0.08
    POSITIVE LOGITS
    ably
    0.13
    AnimationFrame
    0.12
    fully
    0.12
    å·²ä¹ħçļĦ
    0.11
    entially
    0.10
    lessly
    0.10
    é¹ĥ
    0.09
    cripts
    0.09
    å£ij
    0.09
    å·²ä¹ħ
    0.09
    Activations Density 0.022%

    No Known Activations