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. 3-RES-JB
    6. 7735
    Prev
    Next
    INDEX
    Explanations

    function and syntax-related keywords commonly found in programming languages like C++ and libraries like Boost

    oai_token-act-pair · gpt-3.5-turbo

    references to a specific programming library or standard

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    3
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    esville
    -0.86
    staking
    -0.82
    igne
    -0.82
    awaru
    -0.81
    rontal
    -0.79
    cles
    -0.74
    anas
    -0.73
    xon
    -0.71
     Metatron
    -0.70
    undai
    -0.70
    POSITIVE LOGITS
     std
    0.76
    STD
    0.75
    std
    0.72
    Class
    0.72
    ãĥĥãĥī
    0.69
     deduction
    0.69
    OUT
    0.67
    uments
    0.67
    ++
    0.67
    out
    0.67
    Activations Density 0.017%

    No Known Activations