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

    Neuronpedia

    APICircuit 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. 3-RES-JB
    6. 21865
    Prev
    Next
    INDEX
    Explanations

    mentions of specific persons and events in the field of combat sports

    oai_token-act-pair · gpt-3.5-turbo

    references to notable events and accomplishments in sports

    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
    arine
    -0.89
    adobe
    -0.87
    autical
    -0.86
    ibrary
    -0.85
     UNCLASSIFIED
    -0.83
    omsky
    -0.83
    berra
    -0.78
     Gutenberg
    -0.78
    ItemTracker
    -0.78
    earcher
    -0.77
    POSITIVE LOGITS
     rematch
    0.95
     teammate
    0.94
     championship
    0.90
     Showdown
    0.88
     Championship
    0.87
     championships
    0.83
     Wrestle
    0.82
     stopp
    0.82
     MMA
    0.82
     challenger
    0.81
    Activations Density 0.297%

    No Known Activations