© 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. 3-RES-JB
    11. 5822
    Prev
    Next
    INDEX
    Explanations

    superlatives indicating the extreme end of a spectrum

    oai_token-act-pair · gpt-3.5-turbo

    phrases emphasizing the concept of "least" or minimalism

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    3
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    prototype
    -0.77
    gnu
    -0.71
    clerosis
    -0.71
    selves
    -0.70
    axter
    -0.69
    ses
    -0.68
    bj
    -0.67
    dal
    -0.66
    DOM
    -0.63
    halla
    -0.62
    POSITIVE LOGITS
     imaginable
    0.76
     intrusive
    0.69
     squares
    0.65
     conspicuous
    0.63
     ãĤµãĥ¼ãĥĨãĤ£ãĥ¯ãĥ³
    0.63
     toler
    0.61
     credible
    0.60
     visible
    0.60
     soils
    0.60
    ospace
    0.60
    Activations Density 0.013%

    No Known Activations