Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 4-RES-JB
    6. 23495
    Prev
    Next
    INDEX
    Explanations

    NFL player positions such as running backs, wide receivers, defensive backs, and tight ends

    oai_token-act-pair · gpt-3.5-turbo

    positions and roles in a sports context, particularly football

    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
    20439
    -0.74
     Newsp
    -0.70
    beer
    -0.68
    ombat
    -0.68
    imoto
    -0.67
    atom
    -0.66
     Mothers
    -0.64
    zb
    -0.64
     Pictures
    -0.63
     Editors
    -0.62
    POSITIVE LOGITS
    NAME
    0.96
    OUT
    0.86
     guru
    0.86
     Scouting
    0.85
    Coach
    0.85
     linebacker
    0.82
     coach
    0.82
     fullback
    0.77
     cornerback
    0.76
    ACTED
    0.76
    Activations Density 0.074%

    No Known Activations