© 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. 0-RES-JB
    11. 1154
    Prev
    Next
    INDEX
    Explanations

    references to historical figures or entities associated with ancient Rome

    oai_token-act-pair · gpt-3.5-turbo

    references to Roman themes or contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     laun
    -0.76
    FH
    -0.73
     Ellison
    -0.72
     kWh
    -0.71
    APD
    -0.69
    ufact
    -0.69
     DG
    -0.66
    haw
    -0.66
    APP
    -0.66
    heed
    -0.64
    POSITIVE LOGITS
     Roman
    3.71
    Roman
    2.87
     Romans
    2.25
     Rome
    1.72
     Romanian
    1.69
    Greek
    1.53
    roman
    1.48
     Byzantine
    1.46
     Greek
    1.42
     Caesar
    1.40
    Activations Density 0.006%

    No Known Activations