© 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. 1-RES-JB
    11. 15920
    Prev
    Next
    INDEX
    Explanations

    mentions of evil or negative behavior

    oai_token-act-pair · gpt-3.5-turbo

    mentions of the concept of evil

    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
    lov
    -0.86
    PsyNetMessage
    -0.84
    UNCH
    -0.83
    Journal
    -0.75
    Introduced
    -0.74
    Clar
    -0.74
    RESULTS
    -0.74
    Recommend
    -0.74
    Lago
    -0.73
    cially
    -0.72
    POSITIVE LOGITS
     evil
    1.09
     incarn
    1.02
     villain
    0.96
     tyrant
    0.91
     clown
    0.87
     Evil
    0.85
     adversary
    0.85
     genius
    0.85
     mastermind
    0.84
     dece
    0.84
    Activations Density 0.009%

    No Known Activations