© 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. 504
    Prev
    Next
    INDEX
    Explanations

    words related to geographical locations, specifically areas like colonies and provinces

    oai_token-act-pair · gpt-3.5-turbo

    references to colonies and provinces

    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
    statement
    -0.78
    session
    -0.76
     \"
    -0.76
    rette
    -0.75
    osal
    -0.72
     haunt
    -0.71
    gery
    -0.71
    duty
    -0.69
     vehicle
    -0.69
    eway
    -0.68
    POSITIVE LOGITS
     provinces
    2.67
     colonies
    2.12
     factories
    1.80
     divisions
    1.60
     territories
    1.51
     armies
    1.46
     regions
    1.42
     empires
    1.41
     villages
    1.39
     legions
    1.38
    Activations Density 0.023%

    No Known Activations