© 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. 1-RES-JB
    11. 16295
    Prev
    Next
    INDEX
    Explanations

    locations or geographic entities related to territorial threats or military activities

    oai_token-act-pair · gpt-3.5-turbo

    geographical references to specific locations, particularly islands and regions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 1-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    1
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ô
    -0.81
    abus
    -0.80
    chen
    -0.78
    von
    -0.77
    LEY
    -0.74
    aged
    -0.73
    andre
    -0.73
    viation
    -0.73
    fen
    -0.73
    coni
    -0.73
    POSITIVE LOGITS
     Guam
    1.10
     Samoa
    0.88
     Bermuda
    0.86
     Newfoundland
    0.81
     Islands
    0.81
     Greenland
    0.77
     Hurricanes
    0.76
     Labrador
    0.75
     Islanders
    0.74
     Johnston
    0.74
    Activations Density 0.018%

    No Known Activations