© 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. 10-RES-JB
    11. 10785
    Prev
    Next
    INDEX
    Explanations

    words related to malevolence, wickedness, or any form of evil

    oai_token-act-pair · gpt-3.5-turbo

    references to 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 @ 10-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    10
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    PsyNetMessage
    -0.80
    pler
    -0.74
    UNCH
    -0.74
    phrine
    -0.73
    amura
    -0.71
    eding
    -0.70
    ixel
    -0.70
    hene
    -0.69
    aro
    -0.68
    ribe
    -0.68
    POSITIVE LOGITS
     deeds
    1.14
     incarn
    1.11
     deed
    1.02
     mastermind
    0.98
     intentions
    0.95
     spirits
    0.94
     genius
    0.92
     sorce
    0.89
     empire
    0.88
    dies
    0.86
    Activations Density 0.079%

    No Known Activations