© 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. 7833
    Prev
    Next
    INDEX
    Explanations

    phrases related to defining and explaining concepts or ideas

    oai_token-act-pair · gpt-3.5-turbo

    conjunctions, specifically the words "and" and "or."

    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
     Lank
    -0.68
    hoe
    -0.65
     Bundes
    -0.64
     Highland
    -0.63
     Moderate
    -0.62
     Lutheran
    -0.62
     Dual
    -0.60
    igraph
    -0.58
    roth
    -0.58
    ursday
    -0.58
    POSITIVE LOGITS
     how
    0.97
    why
    0.87
    what
    0.77
     why
    0.77
    whether
    0.76
     WHY
    0.76
     HOW
    0.75
     whether
    0.74
     whence
    0.73
     hence
    0.72
    Activations Density 0.187%

    No Known Activations