© 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. 4-RES-JB
    11. 802
    Prev
    Next
    INDEX
    Explanations

    phrases or sentences starting or ending with the word "OK"

    oai_token-act-pair · gpt-3.5-turbo

    sentences or phrases involving hesitation or uncertainty

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

    No Comments

    Negative Logits
    conservancy
    -0.80
    rities
    -0.77
    advertisement
    -0.73
    thood
    -0.71
    reated
    -0.69
    ascript
    -0.69
    ngth
    -0.68
    ription
    -0.67
    plant
    -0.66
    ertation
    -0.64
    POSITIVE LOGITS
    bye
    0.87
    Ô
    0.84
     Yug
    0.73
    ãĤ´ãĥ³
    0.67
     Rik
    0.64
     presum
    0.64
    wy
    0.63
    wo
    0.63
    pie
    0.60
    upid
    0.60
    Activations Density 0.073%

    No Known Activations