© 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. 52609
    Prev
    Next
    INDEX
    Explanations

    message and its content

    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
    odied
    -0.09
    onder
    -0.09
     Misc
    -0.08
    _CHARS
    -0.08
    odd
    -0.08
    ains
    -0.08
    pins
    -0.08
    shi
    -0.08
    apo
    -0.08
    endency
    -0.07
    POSITIVE LOGITS
    \Message
    0.12
    .Message
    0.11
    (message
    0.11
    /message
    0.10
    :message
    0.10
     message
    0.10
    éĢģåĪ°
    0.10
    大çģ«
    0.10
    æīĵå·¥
    0.10
    åĴĮæĶ¯æĮģ
    0.10
    Activations Density 0.011%

    No Known Activations