© 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. 51592
    Prev
    Next
    INDEX
    Explanations

    smash, mash, squashing

    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
    ally
    -0.10
    г
    -0.09
    aits
    -0.08
    verted
    -0.08
    ihan
    -0.08
    ality
    -0.08
    acy
    -0.08
    åīįçŀ»
    -0.08
    rece
    -0.08
    ж
    -0.08
    POSITIVE LOGITS
    tures
    0.14
    intosh
    0.13
    èīĩ
    0.12
    apter
    0.12
    (es
    0.11
    eroon
    0.10
    peare
    0.10
    ingly
    0.10
     mẽ
    0.10
    ola
    0.09
    Activations Density 0.004%

    No Known Activations