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. 4-RES-JB
    6. 21228
    Prev
    Next
    INDEX
    Explanations

    mentions of the name "Rumsfeld"

    oai_token-act-pair · gpt-3.5-turbo

    references to mothers and maternal figures

    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
     resemb
    -0.69
     srf
    -0.65
     departures
    -0.63
    RAW
    -0.63
     Heist
    -0.61
    AUT
    -0.60
     blackout
    -0.60
    Parameters
    -0.59
    XT
    -0.59
     impunity
    -0.58
    POSITIVE LOGITS
    chool
    1.21
    pring
    1.04
    essage
    1.01
    mith
    0.99
    atis
    0.99
    uit
    0.96
    paces
    0.95
    atically
    0.93
    mers
    0.92
    atsu
    0.92
    Activations Density 0.013%

    No Known Activations