© 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. 9-RES-JB
    11. 5468
    Prev
    Next
    INDEX
    Explanations

    information related to tragic events or conflicts, specifically focusing on the number of casualties or deaths

    oai_token-act-pair · gpt-3.5-turbo

    references to casualties or deaths in violent events

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 9-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    9
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    uity
    -0.88
    Redditor
    -0.76
    UTH
    -0.72
    oyer
    -0.71
    Customer
    -0.71
    ENN
    -0.68
    ONEY
    -0.67
    iability
    -0.66
    Truth
    -0.66
    tv
    -0.66
    POSITIVE LOGITS
     gunfire
    1.09
     bombings
    1.06
     shelling
    1.01
     collisions
    1.01
     clashes
    1.00
     bombing
    0.99
     siege
    0.98
     explosions
    0.97
     gunshot
    0.96
     accidents
    0.95
    Activations Density 0.271%

    No Known Activations