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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 17212
    Prev
    Next
    INDEX
    Explanations

    sports champions or individuals holding top titles

    oai_token-act-pair · gpt-3.5-turbo

    references to various champion titles in different sports

    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
    uld
    -0.87
    kefeller
    -0.85
    iday
    -0.76
    iasco
    -0.71
    emetery
    -0.70
    eret
    -0.69
    MpServer
    -0.68
    uters
    -0.66
    gettable
    -0.65
    kinson
    -0.65
    POSITIVE LOGITS
     champion
    0.80
     boxer
    0.79
     contender
    0.77
     challenger
    0.77
     champions
    0.74
     trophy
    0.74
    boards
    0.74
    hip
    0.73
     Champion
    0.71
     laureate
    0.70
    Activations Density 0.009%

    No Known Activations