© 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. 1-RES-JB
    11. 6605
    Prev
    Next
    INDEX
    Explanations

    mentions of individuals with elevated positions in a formal setting such as a police force or government

    oai_token-act-pair · gpt-3.5-turbo

    references to the term "captain."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 1-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    1
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    senal
    -0.84
     inclined
    -0.72
    Downloadha
    -0.71
    FORE
    -0.68
    TI
    -0.65
     sanctions
    -0.65
    DAY
    -0.65
     prone
    -0.64
     unlikely
    -0.63
    20439
    -0.62
    POSITIVE LOGITS
    Capt
    1.23
     Capt
    1.21
    capt
    1.11
    oufl
    1.01
    ivity
    0.88
    ions
    0.87
    ured
    0.84
    cha
    0.83
    alogue
    0.82
    urable
    0.82
    Activations Density 0.005%

    No Known Activations