© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    5. GPT2-Small
    7. Residual Stream
    9. 11-RES-JB
    11. 6925
    Prev
    Next
    INDEX
    Explanations

    descriptions related to physical appearance, particularly focusing on female attributes

    oai_token-act-pair · gpt-3.5-turbo

    expressions related to physical appearance and attractiveness

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 11-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.hook_resid_pre
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    Data Type
    torch.float32
    Hook Point
    blocks.11.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    11
    Activation Function
    relu
    Embeds
    IFrame
    Link
    •Female Descriptors

    No Comments

    Negative Logits
     Gutenberg
    -0.83
     inhibitors
    -0.83
     Firearms
    -0.77
     boycott
    -0.76
     ATF
    -0.76
     citations
    -0.75
     DRM
    -0.74
     authorization
    -0.74
     prohibitions
    -0.73
     ISPs
    -0.72
    POSITIVE LOGITS
     adorable
    1.44
     cute
    1.33
    sweet
    1.32
     beautiful
    1.29
    beaut
    1.29
     handsome
    1.25
     gorgeous
    1.25
     lovely
    1.24
     prett
    1.22
     brun
    1.19
    Activations Density 0.521%

    No Known Activations