© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 20686
    Prev
    Next
    INDEX
    Explanations

    references to the Quran

    oai_token-act-pair · gpt-3.5-turbo

    references to religious texts, specifically the Quran and Torah

    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
    dden
    -0.83
    NetMessage
    -0.81
    yles
    -0.74
    opsy
    -0.69
    ideshow
    -0.68
    izabeth
    -0.67
    alth
    -0.67
    ptive
    -0.66
    emetery
    -0.66
    ptives
    -0.65
    POSITIVE LOGITS
     Quran
    1.42
     Koran
    1.04
     bible
    0.97
     scriptures
    0.94
     Qur
    0.87
    quez
    0.82
     scrolls
    0.81
     manuscript
    0.81
     scripture
    0.80
     manuscripts
    0.79
    Activations Density 0.008%

    No Known Activations