© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    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. 4-RES-JB
    11. 16931
    Prev
    Next
    INDEX
    Explanations

    mentions of the word "Comey"

    oai_token-act-pair · gpt-3.5-turbo

    mentions of James Comey and related discussions surrounding his actions

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

    No Comments

    Negative Logits
    orthodox
    -0.80
    retched
    -0.75
    piring
    -0.72
     Viking
    -0.72
    tainment
    -0.72
    WARD
    -0.71
    cer
    -0.69
     Mond
    -0.69
    tips
    -0.69
    requisite
    -0.68
    POSITIVE LOGITS
     Comey
    0.99
     Clapper
    0.96
     memos
    0.89
     briefings
    0.82
     Emails
    0.82
     dossier
    0.77
     hearings
    0.77
     investigations
    0.76
     testifying
    0.74
    asso
    0.73
    Activations Density 0.023%

    No Known Activations