© 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. 8254
    Prev
    Next
    INDEX
    Explanations

    prawn shrimp

    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
    rics
    -0.13
    itud
    -0.10
    GING
    -0.09
    ramework
    -0.09
    anos
    -0.09
    umin
    -0.09
    umen
    -0.09
    åı¯æĥ³
    -0.09
    andy
    -0.08
    boro
    -0.08
    POSITIVE LOGITS
    åİĿ
    0.09
    -gun
    0.09
    icker
    0.09
    alone
    0.09
    寰
    0.09
     bunch
    0.08
    åĸľ
    0.08
     consolation
    0.08
    że
    0.08
    åĭĥåĭĥ
    0.08
    Activations Density 0.144%

    No Known Activations