© 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. 23601
    Prev
    Next
    INDEX
    Explanations

    mentions of specific buildings or structural features

    oai_token-act-pair · gpt-3.5-turbo

    references to specific buildings, particularly associated with the World Trade Center

    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
    nir
    -0.91
    ican
    -0.84
    sha
    -0.84
    xon
    -0.83
    opard
    -0.80
    romeda
    -0.77
    sweet
    -0.77
    flix
    -0.77
    imum
    -0.74
    icans
    -0.72
    POSITIVE LOGITS
    raper
    1.04
     blocks
    1.02
    Constructed
    1.00
     renovations
    1.00
     occupancy
    1.00
     erected
    0.99
     occupants
    0.94
     buildings
    0.93
    building
    0.93
     demolition
    0.92
    Activations Density 0.058%

    No Known Activations