© 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. 19832
    Prev
    Next
    INDEX
    Explanations

    words ending in 'ort' or 'ogg'

    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
    roma
    -0.10
    èIJ½ä¸ĭ
    -0.10
    åħļçļĦé¢Ĩ导
    -0.09
    èŁł
    -0.08
    姣
    -0.08
    æĺŁè¾°
    -0.08
    uelve
    -0.08
    สาย
    -0.08
    tres
    -0.08
    æŃĩ
    -0.08
    POSITIVE LOGITS
    amy
    0.11
    èµĸ以
    0.09
    -mouth
    0.09
    æį£
    0.09
    mouth
    0.09
    åĺ´
    0.08
    osit
    0.08
    åĦ¿
    0.08
    ä¿ļ
    0.08
    ç³Ĭ
    0.08
    Activations Density 0.158%

    No Known Activations