© 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
    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. 22428
    Prev
    Next
    INDEX
    Explanations

    terms related to government organizations, policies, and regulations

    oai_token-act-pair · gpt-3.5-turbo

    references to anti-corruption, anti-dumping, and anti-fraud measures or organizations

    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
     precincts
    -0.77
     laure
    -0.74
     imag
    -0.74
    ĸļ
    -0.73
     either
    -0.70
     Unch
    -0.70
     Went
    -0.70
     Mons
    -0.69
     pairs
    -0.69
     unaffected
    -0.69
    POSITIVE LOGITS
    corruption
    1.56
    smoking
    1.47
    terrorism
    1.41
    discrimination
    1.40
    gravity
    1.32
    piracy
    1.31
    social
    1.29
    government
    1.28
    terror
    1.28
    terrorist
    1.28
    Activations Density 0.024%

    No Known Activations