© 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. 4-RES-JB
    11. 23687
    Prev
    Next
    INDEX
    Explanations

    references to specific geographical locations, particularly Geneva and The Hague

    oai_token-act-pair · gpt-3.5-turbo

    references to locations related to international law and discussions

    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
    alach
    -0.90
    ramid
    -0.87
    llor
    -0.84
    vich
    -0.77
    adding
    -0.74
    acca
    -0.68
    eding
    -0.68
    intendent
    -0.67
    haw
    -0.67
    transform
    -0.67
    POSITIVE LOGITS
     Geneva
    0.92
     Hague
    0.85
    姫
    0.77
    ventions
    0.74
     seiz
    0.74
     franc
    0.74
     Declaration
    0.73
     Convention
    0.72
     Vienna
    0.71
    ascus
    0.69
    Activations Density 0.015%

    No Known Activations