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

    Neuronpedia

    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. 6-RES-JB
    11. 13505
    Prev
    Next
    INDEX
    Explanations

    phrases related to challenging or reviewing actions by authority figures

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "by" indicating actions or authorship

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 6-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    6
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     resil
    -0.79
    ãĤ¦ãĤ¹
    -0.78
     "$:/
    -0.75
    redits
    -0.70
    stakes
    -0.68
    tenance
    -0.67
    adal
    -0.66
    meat
    -0.65
    URN
    -0.64
    qqa
    -0.64
    POSITIVE LOGITS
     virtue
    1.03
    laws
    0.96
    products
    0.95
    gone
    0.85
    product
    0.78
     omission
    0.77
     proxy
    0.72
     policymakers
    0.69
     whistleblowers
    0.69
     politicians
    0.69
    Activations Density 0.122%

    No Known Activations