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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & 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. 1-RES-JB
    6. 10855
    Prev
    Next
    INDEX
    Explanations

    references to the city of Pyongyang

    oai_token-act-pair · gpt-3.5-turbo

    references to Pyongyang and North Korea

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 1-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    1
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ++++++++++++++++
    -0.88
    title
    -0.81
    brand
    -0.81
    lv
    -0.81
    igor
    -0.79
    RAW
    -0.78
    Torrent
    -0.75
    potion
    -0.74
    TABLE
    -0.74
    RH
    -0.73
    POSITIVE LOGITS
    ongyang
    1.54
     Pyongyang
    1.38
     DPRK
    1.08
     Korea
    0.99
     Seoul
    0.99
    namese
    0.88
     Koreans
    0.87
     Lumpur
    0.85
     Jong
    0.84
    ascus
    0.83
    Activations Density 0.016%

    No Known Activations