© 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. 23828
    Prev
    Next
    INDEX
    Explanations

    rational words and their suffix expansions

    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
    elman
    -0.11
    åĪĩå°Ķ
    -0.09
    ilton
    -0.09
    elve
    -0.09
    лаÑģÑĤ
    -0.09
    èķ¤
    -0.08
    /thumb
    -0.08
    ffer
    -0.08
    ä¸Ģç«Ļ
    -0.08
    ryan
    -0.08
    POSITIVE LOGITS
    istic
    0.14
    ity
    0.14
    ization
    0.11
    主ä¹īèĢħ
    0.11
    izations
    0.10
    سÙħ
    0.10
    istically
    0.10
    主ä¹ī
    0.10
    nels
    0.10
    ITY
    0.10
    Activations Density 0.011%

    No Known Activations