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. 19509
    Prev
    Next
    INDEX
    Explanations

    phrases related to diverse socioeconomic backgrounds

    oai_token-act-pair · gpt-3.5-turbo

    references to diverse social and economic backgrounds

    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
    ught
    -0.82
    ensional
    -0.69
     ther
    -0.68
    cat
    -0.63
     los
    -0.60
     Yankee
    -0.59
    seller
    -0.59
    peak
    -0.59
    ás
    -0.58
    zin
    -0.58
    POSITIVE LOGITS
     backgrounds
    3.26
    Background
    3.04
     background
    2.74
     Background
    2.68
    background
    2.20
     foreground
    1.53
     backdrop
    1.43
     backstory
    1.27
    TextColor
    1.21
    bg
    1.14
    Activations Density 0.018%

    No Known Activations