© 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. 52336
    Prev
    Next
    INDEX
    Explanations

    took part

    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
    éĻĦå±ŀ
    -0.10
    ÙĤÙĨ
    -0.09
    ç²¾åĵģ
    -0.09
    ظر
    -0.09
    onomic
    -0.09
    åķĨçĶ¨
    -0.08
    çīŁ
    -0.08
    æį¢åıĸ
    -0.08
    velopment
    -0.08
    Unmarshaller
    -0.08
    POSITIVE LOGITS
    åĵ§
    0.10
    çĶij
    0.10
    æĪª
    0.10
    æ¹Ľ
    0.10
    sink
    0.09
    å²Ĺ
    0.09
    rows
    0.09
    讳
    0.09
    åĪĹ
    0.09
    è¿Ļ件äºĭ
    0.09
    Activations Density 0.043%

    No Known Activations