© 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. 0-RES-JB
    11. 18976
    Prev
    Next
    INDEX
    Explanations

    mentions of the Devil in texts

    oai_token-act-pair · gpt-3.5-turbo

    references to the concept of the Devil

    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
     gram
    -0.75
     Kyr
    -0.75
    ported
    -0.74
    webkit
    -0.71
    kay
    -0.70
    IA
    -0.70
    rame
    -0.69
    ONSORED
    -0.67
    itures
    -0.66
     redistributed
    -0.64
    POSITIVE LOGITS
     Devil
    3.95
     devil
    2.80
     Devils
    2.69
     devils
    2.49
     Satan
    1.54
     Demons
    1.49
     Lucifer
    1.41
     demons
    1.38
    Dev
    1.36
     Deity
    1.33
    Activations Density 0.014%

    No Known Activations