© 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
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 0-RES-JB
    6. 8389
    Prev
    Next
    INDEX
    Explanations

    phrases related to specific titles or roles, such as "Commissioner"

    oai_token-act-pair · gpt-3.5-turbo

    references to various types of commissioners

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     arts
    -0.75
     kingdom
    -0.72
     quake
    -0.67
     specialize
    -0.66
     soy
    -0.66
     gui
    -0.65
     chick
    -0.64
     yoga
    -0.63
    vu
    -0.63
     presets
    -0.62
    POSITIVE LOGITS
     Commissioner
    3.75
     Commissioners
    2.09
    Commission
    1.98
     commissioner
    1.78
     Superintendent
    1.74
     Commission
    1.53
     Officer
    1.45
     Administrator
    1.45
     Minister
    1.44
     Secretary
    1.40
    Activations Density 0.019%

    No Known Activations