© 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. 22272
    Prev
    Next
    INDEX
    Explanations

    key terms related to political, legal, and current events, such as guidelines, warrants, flights, attacks, and proceedings

    oai_token-act-pair · gpt-3.5-turbo

    terminology related to regulations, directives, and official processes

    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
    odor
    -0.65
    ALLY
    -0.63
    icidal
    -0.63
    lier
    -0.60
    ded
    -0.60
    less
    -0.59
    fac
    -0.59
    rior
    -0.58
    istically
    -0.58
    talk
    -0.58
    POSITIVE LOGITS
    poons
    1.30
    cape
    1.29
    heet
    1.25
    uggest
    1.25
    hip
    1.23
    pring
    1.19
    creen
    1.12
    etting
    1.09
    ourcing
    1.09
    hift
    1.07
    Activations Density 0.638%

    No Known Activations