© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 51597
    Prev
    Next
    INDEX
    Explanations

    names ending in y or words starting with y

    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
    sWith
    -0.17
    ally
    -0.14
    ened
    -0.14
    ively
    -0.12
    ality
    -0.11
    ste
    -0.11
    lessly
    -0.11
    à§į
    -0.11
    ned
    -0.11
    ness
    -0.11
    POSITIVE LOGITS
    outube
    0.18
    ields
    0.18
    esterday
    0.17
    osemite
    0.16
    tics
    0.15
    ramid
    0.15
    ielding
    0.15
    giene
    0.15
    yyyy
    0.15
    psilon
    0.14
    Activations Density 1.457%

    No Known Activations