© 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. 50656
    Prev
    Next
    INDEX
    Explanations

    measurements in cm

    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
    geber
    -0.10
    mund
    -0.09
    verty
    -0.09
    å¤©åĽ½
    -0.09
    edImage
    -0.09
    zilla
    -0.08
    çĦ¶æĺ¯
    -0.08
    edly
    -0.08
    Occurred
    -0.08
    erg
    -0.08
    POSITIVE LOGITS
    ask
    0.09
    两èĢħ
    0.09
    çĿĢæīĭ
    0.09
    èĪĴ
    0.09
    ohen
    0.08
    åĺ²ç¬ij
    0.08
    oko
    0.08
    ä¸Ģéģĵ
    0.08
    éĺľ
    0.08
    edia
    0.08
    Activations Density 0.033%

    No Known Activations