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. 24434
    Prev
    Next
    INDEX
    Explanations

    names of educational institutions related to the suffix "-sburg"

    oai_token-act-pair · gpt-3.5-turbo

    mentions of the city "Pittsburgh."

    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
    Var
    -0.75
     unsustainable
    -0.69
     crystal
    -0.68
    Oracle
    -0.63
     Shadow
    -0.62
     explo
    -0.61
     Catalyst
    -0.59
     view
    -0.59
     spinning
    -0.58
     thin
    -0.58
    POSITIVE LOGITS
    sburg
    5.26
    sburgh
    1.98
    furt
    1.21
    burg
    1.19
    andowski
    1.16
    raltar
    1.15
    sylvania
    1.15
    liga
    1.06
    burgh
    1.03
    lehem
    1.01
    Activations Density 0.017%

    No Known Activations