© 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. 8-RES-JB
    11. 22474
    Prev
    Next
    INDEX
    Explanations

    references to powerful or influential positions in government or organizations

    oai_token-act-pair · gpt-3.5-turbo

    references to high-level authority and corruption in government

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    avia
    -0.72
     Odyssey
    -0.72
    Population
    -0.69
     Reincarn
    -0.69
     Goat
    -0.68
    amaz
    -0.68
     2100
    -0.67
     Neon
    -0.67
     Lucia
    -0.67
    Teen
    -0.65
    POSITIVE LOGITS
     bureaucr
    1.02
     overseeing
    1.01
     bureaucrats
    1.01
     committees
    1.01
     tasked
    0.99
    doms
    0.95
     appointed
    0.95
     vested
    0.92
     ministries
    0.92
     positions
    0.92
    Activations Density 0.424%

    No Known Activations