© 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. 11-RES-JB
    11. 13731
    Prev
    Next
    INDEX
    Explanations

    words related to swapping, exchanging, and trading items or actions

    oai_token-act-pair · gpt-3.5-turbo

    instances of exchanges, trades, or swaps involving items or concepts

    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
    amia
    -0.73
    ĩ
    -0.72
    Loading
    -0.72
    ¯¯
    -0.70
    Ĭ
    -0.70
    ittee
    -0.69
    ¿½
    -0.69
    aston
    -0.67
    âķIJ
    -0.67
    IJ
    -0.67
    POSITIVE LOGITS
     glances
    1.08
     blows
    1.08
     favors
    1.03
     scraps
    0.97
     gifts
    0.88
     fav
    0.86
     punches
    0.85
     metaphors
    0.83
     insults
    0.83
     places
    0.82
    Activations Density 0.069%

    No Known Activations