© 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. 38390
    Prev
    Next
    INDEX
    Explanations

    Matthias or Matthew 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
    åįģä¹Ŀ大
    -0.11
    ä¸Ģç³»åĪĹ
    -0.11
    è··
    -0.10
    AGMA
    -0.10
    atile
    -0.09
    ä¸Ģåı£
    -0.09
    tsx
    -0.09
    çĶľçĶľ
    -0.09
    bé
    -0.09
     Malone
    -0.09
    POSITIVE LOGITS
    èįĶ
    0.08
    eson
    0.08
    actor
    0.08
    lit
    0.08
    udy
    0.08
    acter
    0.08
    bru
    0.08
    hammer
    0.08
    letics
    0.08
    活泼
    0.08
    Activations Density 0.002%

    No Known Activations