© 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. 0-RES-JB
    11. 17988
    Prev
    Next
    INDEX
    Explanations

    mentions of political figures, specifically mayors

    oai_token-act-pair · gpt-3.5-turbo

    references to the position of "mayor."

    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
     Defendant
    -1.02
     Defendants
    -1.01
     Plaintiff
    -0.97
     Members
    -0.95
     Plaint
    -0.90
     Customers
    -0.83
     Membership
    -0.83
     Components
    -0.83
     Companies
    -0.83
     Figures
    -0.82
    POSITIVE LOGITS
     mayor
    2.31
     senator
    2.05
     governor
    1.91
     pope
    1.77
     congressman
    1.66
     president
    1.51
     sheriff
    1.45
     secretary
    1.44
     chancellor
    1.40
     councillor
    1.37
    Activations Density 0.032%

    No Known Activations