© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW 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. 8-RES-JB
    11. 14361
    Prev
    Next
    INDEX
    Explanations

    words related to the soccer team Juventus

    oai_token-act-pair · gpt-3.5-turbo

    mentions of the name "Juventus" and related terms

    oai_token-act-pair · gpt-4o-mini
    Triggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Negative Logits
    ding
    -0.72
    ãĥĥãĥī
    -0.70
    holders
    -0.64
    åº
    -0.63
    raint
    -0.63
    parent
    -0.62
    natal
    -0.59
    etheless
    -0.59
    head
    -0.59
     Haunted
    -0.58
    POSITIVE LOGITS
    xtap
    1.36
    arez
    1.34
    Ju
    1.10
    iced
    1.01
    icer
    1.01
    illet
    0.92
    ju
    0.92
    icy
    0.92
    ice
    0.88
    illard
    0.86
    Activations Density 0.046%

    No Known Activations