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. 3776
    Prev
    Next
    INDEX
    Explanations

    elements related to coding styles or syntax

    oai_token-act-pair · gpt-3.5-turbo

    instances of the term "style."

    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
     arbit
    -0.72
     subcontract
    -0.70
     expend
    -0.68
     aur
    -0.64
     plaintiff
    -0.64
     crater
    -0.63
     cumulative
    -0.63
     residing
    -0.62
     passenger
    -0.62
     comprehend
    -0.61
    POSITIVE LOGITS
    style
    3.93
    Style
    2.15
     style
    2.07
    styles
    1.85
     Style
    1.82
    type
    1.78
    esque
    1.66
    inspired
    1.60
    sty
    1.51
     styles
    1.39
    Activations Density 0.012%

    No Known Activations