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. 8-RES-JB
    6. 6620
    Prev
    Next
    INDEX
    Explanations

    phrases referring to the publication of scientific studies

    oai_token-act-pair · gpt-3.5-turbo

    references to studies published in academic journals

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    chwitz
    -0.83
    HUD
    -0.73
     renting
    -0.71
     leveling
    -0.70
     crou
    -0.69
    chery
    -0.68
     broom
    -0.67
    chers
    -0.67
    Offic
    -0.67
    addon
    -0.67
    POSITIVE LOGITS
     Proceedings
    1.39
     PLoS
    1.32
     Journal
    1.15
    Journal
    1.11
     journal
    1.10
     Pediatrics
    1.08
     Scientific
    1.08
     journals
    1.07
     IEEE
    1.04
     PubMed
    0.99
    Activations Density 0.100%

    No Known Activations