Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    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. 21540
    Prev
    Next
    INDEX
    Explanations

    mentions of specific political figures, particularly mayors

    oai_token-act-pair · gpt-3.5-turbo

    references to mayors

    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
    pill
    -0.77
     pen
    -0.75
    craft
    -0.67
     atom
    -0.67
    corn
    -0.65
    VS
    -0.65
     attachment
    -0.64
    alia
    -0.64
     confuse
    -0.62
    core
    -0.62
    POSITIVE LOGITS
     Mayor
    3.67
    Mayor
    3.32
     mayor
    2.95
     mayors
    2.35
     mayoral
    2.25
     Governor
    1.87
     Municip
    1.70
     councillor
    1.58
     councillors
    1.58
    Council
    1.54
    Activations Density 0.018%

    No Known Activations