Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 0-RES-JB
    6. 10584
    Prev
    Next
    INDEX
    Explanations

    mentions of specific individuals or locations, particularly focusing on Aberdeen and Nicola Sturgeon

    oai_token-act-pair · gpt-3.5-turbo

    names of Scottish political figures and locations

    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
     Liu
    -1.05
     Huma
    -0.94
     Zhao
    -0.84
     Liang
    -0.84
     Xie
    -0.82
     Mek
    -0.82
     jung
    -0.80
     Giov
    -0.78
    <?
    -0.77
     Qiao
    -0.77
    POSITIVE LOGITS
     Scots
    2.73
     Sturgeon
    2.39
     Aberdeen
    2.36
     Edinburgh
    2.26
     Scottish
    2.21
    Scot
    2.19
     Scotland
    2.13
    Scotland
    2.04
     Dund
    1.91
     SNP
    1.88
    Activations Density 0.047%

    No Known Activations