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. 20615
    Prev
    Next
    INDEX
    Explanations

    words related to the medical condition 'ulcer'

    oai_token-act-pair · gpt-3.5-turbo

    variants of the word "ulcer."

    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
     DRAGON
    -0.75
     Pok
    -0.70
     commons
    -0.69
     Stars
    -0.62
    meg
    -0.62
     Cinem
    -0.62
    Come
    -0.61
     Dust
    -0.61
     STAND
    -0.61
    itative
    -0.61
    POSITIVE LOGITS
     ul
    3.89
     Ul
    2.34
    Ul
    1.93
    ul
    1.21
     ult
    1.16
    ullah
    1.03
     cir
    0.96
     sshd
    0.94
     li
    0.94
     athe
    0.90
    Activations Density 0.019%

    No Known Activations