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. 0-RES-JB
    6. 7012
    Prev
    Next
    INDEX
    Explanations

    terms related to MetLife Stadium

    oai_token-act-pair · gpt-3.5-turbo

    references to life, likely focusing on the concept of "life" in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     pretext
    -0.73
     muzzle
    -0.70
     preliminary
    -0.66
     nav
    -0.65
     rout
    -0.65
     backing
    -0.65
     prompting
    -0.64
     foreground
    -0.64
     equ
    -0.63
     balloon
    -0.62
    POSITIVE LOGITS
    Life
    4.02
    life
    2.56
     Life
    2.10
     LIFE
    1.86
     life
    1.54
    Living
    1.52
    lif
    1.49
    Family
    1.39
    Health
    1.29
    Surv
    1.25
    Activations Density 0.017%

    No Known Activations