© 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. 11-RES-JB
    11. 6814
    Prev
    Next
    INDEX
    Explanations

    words related to accidents and incidents

    oai_token-act-pair · gpt-3.5-turbo

    references to accidents and their consequences

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 11-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    11
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    wick
    -0.76
    ringe
    -0.73
    tsky
    -0.69
    hett
    -0.68
    HCR
    -0.68
    ongyang
    -0.68
    endas
    -0.66
    Intern
    -0.66
    rica
    -0.66
    ramids
    -0.65
    POSITIVE LOGITS
     accident
    1.02
    worthiness
    1.00
     hazard
    0.98
     involving
    0.97
     occurring
    0.88
     occ
    0.88
     causation
    0.87
     fatalities
    0.86
     Course
    0.85
     Investigation
    0.85
    Activations Density 0.062%

    No Known Activations