© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW 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. 8-RES-JB
    11. 14264
    Prev
    Next
    INDEX
    Explanations

    terms associated with significant violence, conflict, and humanitarian crises

    oai_token-act-pair · gpt-4o-mini
    Triggered by @bot

    phrases related to severity or intensity

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Negative Logits
     buddies
    -0.89
     wisely
    -0.87
     mates
    -0.79
    mates
    -0.78
     Gabe
    -0.75
    */(
    -0.74
     keeper
    -0.71
     Handler
    -0.70
    Favorite
    -0.69
     Investigator
    -0.68
    POSITIVE LOGITS
     bloodshed
    1.42
     crackdown
    1.26
     upheaval
    1.25
     violence
    1.22
     violent
    1.21
     carnage
    1.20
     unrest
    1.19
     protests
    1.16
     clashes
    1.16
     deterioration
    1.16
    Activations Density 0.529%

    No Known Activations