© 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. 18205
    Prev
    Next
    INDEX
    Explanations

    references to lengths or distances expressed in yards

    oai_token-act-pair · gpt-3.5-turbo

    references to yardage measurements in sports contexts

    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
    atively
    -0.72
    isen
    -0.72
     plur
    -0.72
     anat
    -0.68
    ipper
    -0.64
     consulted
    -0.62
    hran
    -0.61
    yu
    -0.61
    izoph
    -0.61
    regon
    -0.61
    POSITIVE LOGITS
    yard
    0.90
     touchdown
    0.84
    yards
    0.81
    ometers
    0.81
     Kare
    0.76
     touchdowns
    0.75
    birds
    0.75
     yards
    0.74
     cornerback
    0.71
     coach
    0.69
    Activations Density 0.009%

    No Known Activations