© 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. 0-RES-JB
    11. 8368
    Prev
    Next
    INDEX
    Explanations

    words related to spacesuits and space travel

    oai_token-act-pair · gpt-3.5-turbo

    references to suits or suit-like items

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Bates
    -0.68
     Sid
    -0.62
     Farms
    -0.62
     Candidate
    -0.61
     rounded
    -0.59
     feeds
    -0.58
     draft
    -0.58
     farms
    -0.57
     marks
    -0.56
     Wicked
    -0.56
    POSITIVE LOGITS
    uit
    4.88
    uits
    3.80
    uitous
    1.37
    ued
    1.27
    uum
    1.24
    uin
    1.24
    uing
    1.21
    uity
    1.17
    hirt
    1.17
    uate
    1.14
    Activations Density 0.010%

    No Known Activations