© 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. 4-RES-JB
    11. 19689
    Prev
    Next
    INDEX
    Explanations

    terms related to geographical locations, governmental bodies, and political events

    oai_token-act-pair · gpt-3.5-turbo

    references to the United States and the United Kingdom in a political context

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

    No Comments

    Negative Logits
    ructure
    -0.84
    epad
    -0.83
    iest
    -0.79
    bsite
    -0.77
    agame
    -0.76
    oscope
    -0.74
    ossibility
    -0.71
    ultimate
    -0.69
    iom
    -0.69
    anse
    -0.69
    POSITIVE LOGITS
     Affairs
    0.81
    Leaks
    0.77
     Counsel
    0.75
     officials
    0.72
     Advis
    0.72
     Column
    0.71
     counterterrorism
    0.71
     briefings
    0.71
     diplomats
    0.70
     veterans
    0.70
    Activations Density 0.509%

    No Known Activations