© 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. 7941
    Prev
    Next
    INDEX
    Explanations

    am/om endings leading to suffixes

    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
    常å§Ķ
    -0.18
    *******/↵
    -0.13
    ermen
    -0.11
    inch
    -0.11
    اÙħر
    -0.11
    ÙĨا
    -0.11
    ÙĨÙĬ
    -0.11
    ermo
    -0.10
    minster
    -0.10
    omore
    -0.10
    POSITIVE LOGITS
    atically
    0.22
    perature
    0.19
    ming
    0.19
    soever
    0.15
    atics
    0.13
    bers
    0.13
    oney
    0.13
    ichael
    0.13
    ulator
    0.13
    orrow
    0.13
    Activations Density 2.010%

    No Known Activations