© 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. 37125
    Prev
    Next
    INDEX
    Explanations

    street names

    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
    ìĦĿ
    -0.10
    odic
    -0.10
    er
    -0.09
    quat
    -0.09
    itat
    -0.08
    pec
    -0.08
    erer
    -0.08
    éľīç´ł
    -0.08
    èµģ
    -0.08
    ium
    -0.08
    POSITIVE LOGITS
    bars
    0.09
    ä¸ĵ项èµĦéĩij
    0.09
    å··
    0.08
     اÙĦدر
    0.08
    ä¸Ģ度
    0.08
    åĿĬ
    0.08
    arity
    0.08
    gear
    0.08
    reme
    0.08
     Cooler
    0.08
    Activations Density 0.014%

    No Known Activations