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

    Neuronpedia

    APIAssistant AxisNEWCircuit 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. 15202
    Prev
    Next
    INDEX
    Explanations

    information related to intelligence gathering, surveillance, potential threats, and cyber attacks

    oai_token-act-pair · gpt-3.5-turbo

    mentions of surveillance and privacy issues

    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
     Towns
    -0.80
     Daly
    -0.74
    ciation
    -0.74
     Brisbane
    -0.72
     Volvo
    -0.72
     evict
    -0.69
     Glasgow
    -0.67
     Calgary
    -0.67
     Arcade
    -0.67
     Notting
    -0.66
    POSITIVE LOGITS
     espionage
    1.56
    ionage
    1.55
     spies
    1.38
     intelligence
    1.35
    intelligence
    1.30
     whistleblowers
    1.30
    CLASSIFIED
    1.29
     cybersecurity
    1.28
     adversaries
    1.27
     counterterrorism
    1.26
    Activations Density 0.516%

    No Known Activations