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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 4-RES-JB
    11. 18639
    Prev
    Next
    INDEX
    Explanations

    phrases related to legislative processes, particularly actions and discussions taking place on the floor of a governing body

    oai_token-act-pair · gpt-3.5-turbo

    references to legislative proceedings or discussions occurring on the floor of a legislative body

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

    No Comments

    Negative Logits
    elson
    -0.74
    ather
    -0.71
     Alonso
    -0.70
    eday
    -0.67
    enegger
    -0.66
    OVA
    -0.63
     Occ
    -0.63
    ortium
    -0.63
    omers
    -0.63
     Brav
    -0.62
    POSITIVE LOGITS
     floor
    1.11
    boards
    0.92
     Floor
    0.91
    floor
    0.88
     mats
    0.87
    room
    0.87
     floors
    0.85
     ceiling
    0.81
    nuts
    0.81
    parts
    0.80
    Activations Density 0.008%

    No Known Activations