© 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. 7-RES-JB
    11. 22961
    Prev
    Next
    INDEX
    Explanations

    terms related to scientific research and development

    oai_token-act-pair · gpt-3.5-turbo

    references to research activities and initiatives

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 7-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    7
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     lash
    -0.62
    ĪĴ
    -0.62
     rapp
    -0.62
     Brow
    -0.61
     chopped
    -0.60
    emonic
    -0.60
     Blaze
    -0.59
     jerk
    -0.59
    cakes
    -0.58
     Rule
    -0.58
    POSITIVE LOGITS
     labs
    0.86
     laboratories
    0.86
     scientist
    0.85
     instit
    0.82
     conducted
    0.82
     undertaken
    0.80
     assistants
    0.80
     Scientist
    0.78
    Researchers
    0.76
     findings
    0.76
    Activations Density 0.051%

    No Known Activations