© 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. 4-RES-JB
    11. 14474
    Prev
    Next
    INDEX
    Explanations

    words related to catastrophic events or emergencies

    oai_token-act-pair · gpt-3.5-turbo

    mentions of disasters and their impacts

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

    No Comments

    Negative Logits
    ells
    -0.87
    eton
    -0.84
    kaya
    -0.81
    Ģ
    -0.79
    ularity
    -0.78
    orney
    -0.77
    rab
    -0.76
    clud
    -0.74
    appa
    -0.74
    skirts
    -0.72
    POSITIVE LOGITS
     catast
    1.07
     engulf
    0.97
     Disaster
    0.88
     devast
    0.86
     disaster
    0.86
     plag
    0.83
     Survivor
    0.82
     Outbreak
    0.81
     Katrina
    0.80
     catastrophe
    0.80
    Activations Density 0.018%

    No Known Activations