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. 9-RES-JB
    6. 1286
    Prev
    Next
    INDEX
    Explanations

    names of people, positions, and organizations related to the FBI and other government agencies

    oai_token-act-pair · gpt-3.5-turbo

    references to FBI leadership and related investigations

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 9-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    9
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     WWF
    -0.87
     Parables
    -0.86
     Jagu
    -0.79
     relegation
    -0.78
     Municip
    -0.78
     broch
    -0.78
     Winnipeg
    -0.77
     Indust
    -0.76
     billboards
    -0.76
     Kamp
    -0.76
    POSITIVE LOGITS
     Comey
    1.24
    FBI
    1.17
     Clapper
    1.02
     subpoen
    1.02
     DOJ
    1.01
     FBI
    1.00
     testifying
    0.98
     Rosenstein
    0.94
     Mattis
    0.94
     overseeing
    0.93
    Activations Density 0.434%

    No Known Activations