Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 11-RES-JB
    6. 7951
    Prev
    Next
    INDEX
    Explanations

    references to terrorist groups, specifically focusing on Al Qaeda

    oai_token-act-pair · gpt-3.5-turbo

    mentions of the organization Al-Qaeda and related entities

    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
     sterling
    -0.63
     enchanted
    -0.60
     entirely
    -0.57
     goodbye
    -0.57
     revolving
    -0.55
     Twain
    -0.55
    SHIP
    -0.55
     permanently
    -0.55
     guiActiveUnfocused
    -0.54
    ALLY
    -0.53
    POSITIVE LOGITS
     Qaeda
    1.42
    Qaeda
    1.14
    Qaida
    1.08
    oof
    0.98
    leys
    0.97
    chemy
    0.96
    ighting
    0.95
     Jazeera
    0.94
    ihadi
    0.93
    ibaba
    0.92
    Activations Density 0.022%

    No Known Activations