© 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 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. 3-RES-JB
    11. 4071
    Prev
    Next
    INDEX
    Explanations

    numerical patterns related to sports scores or statistics with a focus on 3-point actions

    oai_token-act-pair · gpt-3.5-turbo

    numerical values related to sports statistics

    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
    ulhu
    -0.86
     photoc
    -0.74
    esson
    -0.72
     Moj
    -0.71
     moder
    -0.69
     Dickinson
    -0.68
     ®
    -0.68
    aii
    -0.66
     SX
    -0.66
     âĺ
    -0.66
    POSITIVE LOGITS
    season
    1.16
    sized
    1.04
    game
    1.04
    eligible
    1.03
    winning
    1.01
    football
    1.00
    NFL
    0.99
    player
    0.97
    scoring
    0.97
    team
    0.97
    Activations Density 0.171%

    No Known Activations