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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI 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. 8-RES-JB
    11. 1200
    Prev
    Next
    INDEX
    Explanations

    dates and numerical patterns

    oai_token-act-pair · gpt-3.5-turbo

    structured numerical data and ratings

    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
     matt
    -0.80
     earthquakes
    -0.69
     tsunami
    -0.67
     Amar
    -0.64
     ko
    -0.63
     transports
    -0.63
     earthquake
    -0.63
     è£ıè
    -0.61
     seism
    -0.61
     HF
    -0.60
    POSITIVE LOGITS
     guessed
    0.73
    ł
    0.70
    Ĥİ
    0.65
    iltr
    0.65
    ¼
    0.64
    ¢
    0.64
    º
    0.64
    ŀ
    0.64
    ©¶æ
    0.63
    µ
    0.62
    Activations Density 0.145%

    No Known Activations