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. 23852
    Prev
    Next
    INDEX
    Explanations

    mentions or references to members of a specific group or community

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "Members."

    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
     jugg
    -0.82
     aven
    -0.76
     port
    -0.68
     center
    -0.68
     king
    -0.66
     whis
    -0.65
     mill
    -0.65
     solid
    -0.64
     god
    -0.64
     heel
    -0.64
    POSITIVE LOGITS
     Members
    3.37
    Members
    2.23
     Member
    2.21
     Membership
    1.98
    Member
    1.81
    members
    1.62
     Representatives
    1.49
     Guests
    1.49
     Legislation
    1.43
     Parties
    1.42
    Activations Density 0.016%

    No Known Activations