© 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. 2-RES-JB
    11. 7858
    Prev
    Next
    INDEX
    Explanations

    phrases related to specific individuals in political and military contexts, specifically focusing on the names "David Petraeus" and "James Mattis."

    oai_token-act-pair · gpt-3.5-turbo

    mentions of key political and military figures, particularly David Petraeus

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 2-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    2
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    semble
    -0.70
     Glob
    -0.69
    unal
    -0.69
    ership
    -0.69
    nings
    -0.65
    ruciating
    -0.64
    izzy
    -0.64
    WIND
    -0.63
    ashes
    -0.63
    Ķ
    -0.63
    POSITIVE LOGITS
     Petraeus
    1.23
    ilon
    0.90
    arnaev
    0.84
    oleon
    0.80
    aeda
    0.78
    espie
    0.75
    feld
    0.73
    ileaks
    0.72
    ghazi
    0.69
    olitics
    0.67
    Activations Density 0.007%

    No Known Activations