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

    Neuronpedia

    APISteerSAE EvalsBlog/PodcastNEWSlackPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlog/PodcastGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 5-RES-JB
    6. 8027
    Prev
    Next
    INDEX
    Explanations

    specific parts of a building, such as "roof," "rooftops," and "roofing materials."

    oai_token-act-pair · gpt-3.5-turbo

    repeated mentions of the word "roof" in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 5-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    5
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     MacArthur
    -0.79
    avez
    -0.73
     Sacrament
    -0.72
     Lauder
    -0.71
    ISTER
    -0.66
     Schumer
    -0.65
    amate
    -0.65
    asure
    -0.63
    bourg
    -0.62
    imens
    -0.62
    POSITIVE LOGITS
     roof
    1.00
    deck
    0.96
     rack
    0.86
    ilion
    0.86
     terr
    0.84
    miah
    0.84
    heed
    0.83
     tops
    0.81
    stairs
    0.81
     loft
    0.81
    Activations Density 0.012%

    No Known Activations