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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 8-RES-JB
    11. 6185
    Prev
    Next
    INDEX
    Explanations

    the name "Han" followed by any other token, with different activation levels based on specific contexts

    oai_token-act-pair · gpt-3.5-turbo

    mentions of the name "Han."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Thumbnails
    -0.68
     destro
    -0.65
    ODUCT
    -0.65
    cipline
    -0.64
    tsky
    -0.63
     Colossus
    -0.63
     IMAGES
    -0.62
    utics
    -0.62
    llan
    -0.61
    Downloadha
    -0.61
    POSITIVE LOGITS
    auer
    1.12
    ning
    1.04
    nington
    0.99
    lon
    0.99
    wei
    0.98
    uman
    0.98
     Solo
    0.95
    wal
    0.88
    ako
    0.86
    ergy
    0.86
    Activations Density 0.027%

    No Known Activations