© 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. 20219
    Prev
    Next
    INDEX
    Explanations

    phrases indicating concession or acknowledgment, such as "Granted," "Admittedly," and "Granted, but."

    oai_token-act-pair · gpt-3.5-turbo

    phrases that indicate concession or acknowledgment

    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
    que
    -0.84
    leading
    -0.76
    natureconservancy
    -0.74
    taboola
    -0.73
    leaders
    -0.71
    cum
    -0.69
    QL
    -0.69
    mented
    -0.68
    urated
    -0.68
    bin
    -0.66
    POSITIVE LOGITS
    imaru
    0.82
    wagen
    0.73
    lihood
    0.71
     hindsight
    0.70
    terday
    0.65
    ardon
    0.64
     Mons
    0.64
     Odyssey
    0.63
    adays
    0.63
     admittedly
    0.63
    Activations Density 0.014%

    No Known Activations