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. 11-RES-JB
    6. 2881
    Prev
    Next
    INDEX
    Explanations

    mentions of repealing and making changes to a specific healthcare policy or law

    oai_token-act-pair · gpt-3.5-turbo

    phrases related to the concept of repealing laws, specifically healthcare legislation

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 11-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    11
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    eele
    -0.70
    pring
    -0.65
     liner
    -0.65
    liga
    -0.64
    coni
    -0.61
     enthusi
    -0.60
     anecd
    -0.59
     guiActiveUnfocused
    -0.59
     lapt
    -0.59
    Laughs
    -0.58
    POSITIVE LOGITS
     Obamacare
    1.46
     ObamaCare
    1.33
    amacare
    1.27
     Dodd
    1.08
     ACA
    1.05
     Affordable
    0.98
     DACA
    0.95
     Roe
    0.92
    Ob
    0.90
    ACA
    0.89
    Activations Density 0.070%

    No Known Activations