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. 1-RES-JB
    6. 17249
    Prev
    Next
    INDEX
    Explanations

    keywords related to data views or visualizations

    oai_token-act-pair · gpt-3.5-turbo

    references to UI components in a programming context

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 1-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    1
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    cial
    -0.82
     deaf
    -0.81
    vernment
    -0.79
     whistle
    -0.78
    bably
    -0.77
    cffff
    -0.76
     quieter
    -0.72
     seal
    -0.70
     cold
    -0.69
    ursed
    -0.68
    POSITIVE LOGITS
    View
    1.02
    ById
    0.99
     View
    0.98
    edIn
    0.93
    Model
    0.92
     largeDownload
    0.91
    eous
    0.87
    points
    0.87
    finder
    0.86
    ports
    0.86
    Activations Density 0.007%

    No Known Activations