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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 22278
    Prev
    Next
    INDEX
    Explanations

    phrases related to organizations or institutions, specifically those referred to as agencies

    oai_token-act-pair · gpt-3.5-turbo

    mentions of various agencies, particularly governmental or organizational bodies

    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
     Lect
    -0.64
     backwards
    -0.64
    Spread
    -0.64
    hess
    -0.63
    ustom
    -0.62
    linux
    -0.61
    introdu
    -0.60
    ftime
    -0.60
     Pik
    -0.60
     kne
    -0.60
    POSITIVE LOGITS
     agency
    3.78
     Agency
    2.67
    agency
    2.51
     agencies
    2.49
    gency
    1.73
     bureau
    1.72
     agent
    1.65
     agents
    1.51
    agent
    1.48
     department
    1.37
    Activations Density 0.017%

    No Known Activations