Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 0-RES-JB
    6. 17345
    Prev
    Next
    INDEX
    Explanations

    terms related to the Islamic State

    oai_token-act-pair · gpt-3.5-turbo

    references to Islam and Islamic groups

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    uckle
    -0.71
    CG
    -0.71
    wagon
    -0.70
    CV
    -0.68
     Crusher
    -0.68
    DK
    -0.67
     sled
    -0.66
     tether
    -0.66
     Craw
    -0.64
     CG
    -0.63
    POSITIVE LOGITS
     Islamic
    3.66
    Islamic
    3.25
     Islamist
    2.38
     Islam
    2.31
     Muslim
    2.26
    Islam
    2.12
     Islamists
    2.05
     Muslims
    2.01
     Sharia
    2.01
    Muslim
    1.91
    Activations Density 0.017%

    No Known Activations