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

    Neuronpedia

    Jacobian LensNEW
    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. 12699
    Prev
    Next
    INDEX
    Explanations

    references to the concept of "Hell."

    oai_token-act-pair · gpt-3.5-turbo

    references to "Hell."

    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
    apo
    -0.69
    tips
    -0.68
    voice
    -0.68
    jiang
    -0.67
    webkit
    -0.66
    ndra
    -0.66
     jog
    -0.65
    rade
    -0.64
    Reference
    -0.63
     CPI
    -0.62
    POSITIVE LOGITS
     Hell
    3.83
    Hell
    2.77
     hell
    2.33
     HELL
    2.28
     Heck
    1.69
     Heaven
    1.60
     Hades
    1.53
     Purg
    1.50
     Inferno
    1.45
     Fuck
    1.37
    Activations Density 0.014%

    No Known Activations