© 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. 4-RES-JB
    11. 10776
    Prev
    Next
    INDEX
    Explanations

    Roman numerals, specifically focusing on the numbers VII and VIII

    oai_token-act-pair · gpt-3.5-turbo

    references to historical figures, specifically kings and monarchs

    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
    feed
    -0.90
    href
    -0.89
    bies
    -0.83
    м
    -0.78
    boards
    -0.77
    bery
    -0.76
    Ú
    -0.76
    rill
    -0.75
    bie
    -0.75
    ãĤ±
    -0.73
    POSITIVE LOGITS
    III
    1.18
    II
    0.93
     Jinping
    0.88
     VIII
    0.80
    XX
    0.77
    avorite
    0.76
     III
    0.74
     XII
    0.74
    VII
    0.73
     VII
    0.73
    Activations Density 0.014%

    No Known Activations