© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 19682
    Prev
    Next
    INDEX
    Explanations

    Dev or dev followed by word parts/words

    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
    pped
    -0.10
    裳
    -0.09
    ãģĹãģ¾ãģĨ
    -0.09
    /or
    -0.09
    веÑĢÑħ
    -0.09
    ings
    -0.09
    Vectorizer
    -0.09
    _VAR
    -0.08
    cion
    -0.08
    hausen
    -0.08
    POSITIVE LOGITS
    iations
    0.13
    uxe
    0.12
    iation
    0.12
    ends
    0.11
    iates
    0.10
    èĩ¾
    0.10
    ils
    0.10
    ience
    0.10
    ign
    0.10
    uelve
    0.10
    Activations Density 0.032%

    No Known Activations