© 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. 17089
    Prev
    Next
    INDEX
    Explanations

    adjectives followed by nouns indicating negative characteristics

    oai_token-act-pair · gpt-3.5-turbo

    articles and adjectives describing people negatively

    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
    Not in Any Lists

    No Comments

    Negative Logits
    ARB
    -0.72
    encies
    -0.71
    aris
    -0.71
    eton
    -0.71
    iversary
    -0.71
    Items
    -0.70
    articles
    -0.70
    Access
    -0.70
    acion
    -0.69
    onds
    -0.69
    POSITIVE LOGITS
     liar
    1.35
     hypocr
    1.29
     believer
    1.14
     bigot
    1.13
     traitor
    1.12
     coward
    1.11
     genius
    1.11
     patriot
    1.10
     tyrant
    1.09
     visionary
    1.06
    Activations Density 0.170%

    No Known Activations