© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    5. GPT2-Small
    7. Residual Stream
    9. 4-RES-JB
    11. 10167
    Prev
    Next
    INDEX
    Explanations

    mentions of the word "Chernobyl"

    oai_token-act-pair · gpt-3.5-turbo

    references to the Chernobyl disaster and related topics

    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
    fol
    -0.77
    flo
    -0.74
    orb
    -0.70
    board
    -0.69
    lease
    -0.69
    Construct
    -0.69
    ERC
    -0.66
    Spread
    -0.66
    bes
    -0.65
    ITS
    -0.65
    POSITIVE LOGITS
    okers
    0.87
     Tobacco
    0.87
     smoker
    0.85
     smoked
    0.80
    urrent
    0.79
    oky
    0.78
    ategory
    0.73
     tobacco
    0.72
    ²¾
    0.72
    creen
    0.72
    Activations Density 0.031%

    No Known Activations