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. 0-RES-JB
    6. 17792
    Prev
    Next
    INDEX
    Explanations

    instances where something is physically wrapped

    oai_token-act-pair · gpt-3.5-turbo

    occurrences of the word "wrapped" and its variations 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
    edia
    -0.76
     Rate
    -0.69
    peak
    -0.68
     Cosponsors
    -0.66
     Correction
    -0.64
    iens
    -0.63
    edom
    -0.61
    auri
    -0.61
     mete
    -0.61
    illy
    -0.61
    POSITIVE LOGITS
     wrapped
    3.50
     wrap
    2.21
     wrapping
    2.20
     wraps
    2.02
    wra
    1.85
     Wra
    1.69
     Wrap
    1.65
     wrapper
    1.59
     draped
    1.56
    wrap
    1.49
    Activations Density 0.015%

    No Known Activations