© 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. 34455
    Prev
    Next
    INDEX
    Explanations

    house rules

    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
    shaw
    -0.13
    ships
    -0.12
    ìĤ¬íķŃ
    -0.11
    à¹Į
    -0.11
    ãģĦãģŁãĤī
    -0.10
    lix
    -0.10
    ship
    -0.10
    sh
    -0.10
    thal
    -0.09
    shift
    -0.09
    POSITIVE LOGITS
    mates
    0.13
    keeping
    0.13
    maid
    0.12
    oft
    0.12
    brew
    0.11
    ier
    0.11
    boat
    0.11
    kü
    0.11
    mate
    0.10
    holds
    0.10
    Activations Density 0.040%

    No Known Activations