© 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. 48855
    Prev
    Next
    INDEX
    Explanations

    input parameter or string

    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
    å¤ĸåĩº
    -0.12
    åĩ«
    -0.11
    adier
    -0.10
    åīįåı°
    -0.10
    åĩºæ¸¸
    -0.10
    åĩºåľº
    -0.10
     вÑĭÑĢажен
    -0.10
    enal
    -0.09
    ãģ¶ãĤĬ
    -0.09
    inn
    -0.09
    POSITIVE LOGITS
    ted
    0.18
    /output
    0.15
    ters
    0.13
    ting
    0.12
    MethodManager
    0.12
    -output
    0.12
    ners
    0.11
    TING
    0.11
    .LA
    0.10
    prising
    0.10
    Activations Density 0.273%

    No Known Activations