© 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. 8-RES-JB
    11. 6170
    Prev
    Next
    INDEX
    Explanations

    mentions of terrorist groups and their activities

    oai_token-act-pair · gpt-3.5-turbo

    references to terrorist groups and their activities

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    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
    Not in Any Lists

    No Comments

    Negative Logits
     Canal
    -0.83
    anan
    -0.72
    ikk
    -0.70
    Scot
    -0.70
    é¾įå
    -0.69
     Seah
    -0.69
     Rhod
    -0.69
     Startup
    -0.67
    ecd
    -0.67
    uci
    -0.66
    POSITIVE LOGITS
     attacking
    1.00
     impunity
    0.94
     menacing
    0.93
     prow
    0.92
     exploiting
    0.90
     rampage
    0.90
     targeting
    0.90
     murdering
    0.90
     explosives
    0.89
     undet
    0.89
    Activations Density 0.893%

    No Known Activations