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. 11-RES-JB
    6. 8131
    Prev
    Next
    INDEX
    Explanations

    word sequences that represent the structure of a letter

    oai_token-act-pair · gpt-3.5-turbo

    references to letters and written correspondence

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 11-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    11
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     booths
    -0.82
    Course
    -0.76
     documentaries
    -0.75
    ¥µ
    -0.72
    Stage
    -0.72
    oggles
    -0.72
     exhibitions
    -0.72
    Plex
    -0.69
    "},"
    -0.68
     Races
    -0.67
    POSITIVE LOGITS
     letter
    1.76
     letters
    1.46
     dated
    1.44
     handwritten
    1.39
    Dear
    1.38
    letter
    1.36
     forwarded
    1.33
     mailed
    1.32
     Letter
    1.29
     correspondence
    1.28
    Activations Density 0.311%

    No Known Activations