© 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. 3-RES-JB
    11. 16174
    Prev
    Next
    INDEX
    Explanations

    words related to different parts of a structure, particularly decks and banks

    oai_token-act-pair · gpt-3.5-turbo

    references to structures or areas related to decks and banks

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    3
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    IQ
    -0.74
     fingerprints
    -0.71
     finance
    -0.68
    ItemThumbnailImage
    -0.67
     qualitative
    -0.65
     dearly
    -0.64
     superpower
    -0.62
    ĻĤ
    -0.61
    UGC
    -0.61
    Money
    -0.61
    POSITIVE LOGITS
    bridge
    0.94
     ridge
    0.85
     rim
    0.80
     shaft
    0.80
    mast
    0.79
     roof
    0.78
     overlooking
    0.78
     mast
    0.78
     Pier
    0.77
    south
    0.76
    Activations Density 0.259%

    No Known Activations