© 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. 10-RES-JB
    11. 9964
    Prev
    Next
    INDEX
    Explanations

    mentions of jail or jail-related terms

    oai_token-act-pair · gpt-3.5-turbo

    references to jail and incarceration

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    10
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Nare
    -0.75
    uyomi
    -0.72
     Investors
    -0.66
     Illum
    -0.65
     Scotia
    -0.64
    Downloadha
    -0.64
     Rossi
    -0.63
    ãĥ¤
    -0.61
     Voters
    -0.60
     Lynd
    -0.58
    POSITIVE LOGITS
    break
    1.22
    broken
    1.21
    house
    1.14
    breaks
    1.12
    breaking
    1.02
    bird
    0.98
    breakers
    0.98
    rooms
    0.94
    birds
    0.90
    breaker
    0.89
    Activations Density 0.031%

    No Known Activations