© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 48241
    Prev
    Next
    INDEX
    Explanations

    is called or described as

    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
    ufs
    -0.10
    è¡ħ
    -0.09
    主è¦ģæľī
    -0.08
    æĤŁ
    -0.08
    uf
    -0.08
     matters
    -0.08
    shaw
    -0.08
    åĻĹ
    -0.08
    å°±æĺ¯è¦ģ
    -0.08
    .defaults
    -0.08
    POSITIVE LOGITS
    称ä¹ĭ为
    0.26
    称ä¹ĭ
    0.24
    称为
    0.21
    称
    0.20
    稱
    0.19
    è§Ĩ为
    0.17
    称åij¼
    0.17
    被称为
    0.16
    åı«åģļ
    0.16
    èĩªç§°
    0.16
    Activations Density 0.148%

    No Known Activations