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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 8-RES-JB
    11. 23905
    Prev
    Next
    INDEX
    Explanations

    text snippets related to captions or excerpts in various contexts, including media, reports, and social media posts

    oai_token-act-pair · gpt-3.5-turbo

    references to excerpts and benchmarks, particularly in the context of captions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    aders
    -0.79
    uana
    -0.75
    mos
    -0.69
    cius
    -0.69
    ULTS
    -0.67
    udder
    -0.67
    ggle
    -0.65
    iverpool
    -0.65
    erno
    -0.64
    asion
    -0.63
    POSITIVE LOGITS
    ed
    2.01
    ing
    1.70
    edly
    1.58
    eers
    1.04
    eering
    1.02
    eer
    1.01
    edIn
    1.00
    es
    0.95
    er
    0.94
    eting
    0.93
    Activations Density 0.166%

    No Known Activations