© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    5. GPT2-Small
    7. Residual Stream
    9. 11-RES-JB
    11. 22538
    Prev
    Next
    INDEX
    Explanations

    Roman numerals indicating volumes or parts in a series

    oai_token-act-pair · gpt-3.5-turbo

    components related to volume or parts within a structured content, indicating organization or segmentation

    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
     confir
    -0.86
     ric
    -0.81
     independ
    -0.77
    kefeller
    -0.76
    ickr
    -0.75
     endors
    -0.73
     sling
    -0.70
     laun
    -0.70
     cous
    -0.69
     wom
    -0.68
    POSITIVE LOGITS
     Chapter
    1.91
    Chapter
    1.67
     Chapters
    1.51
     Part
    1.50
     chapter
    1.42
    Section
    1.37
     Appendix
    1.36
     Volume
    1.34
     Episode
    1.34
     Section
    1.32
    Activations Density 0.213%

    No Known Activations