© 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. 1-RES-JB
    11. 12075
    Prev
    Next
    INDEX
    Explanations

    mentions of the word "Pope" along with some specific contexts, such as visits or mentions of names like Benedict

    oai_token-act-pair · gpt-3.5-turbo

    mentions of the Pope

    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
    ript
    -0.82
    yrinth
    -0.80
    erry
    -0.73
    mental
    -0.73
    rg
    -0.71
    ties
    -0.71
    CAP
    -0.70
    é¾
    -0.70
    uld
    -0.69
    awks
    -0.68
    POSITIVE LOGITS
     Francis
    1.11
     Pope
    1.08
    Pope
    1.02
     pont
    0.90
     pope
    0.87
     Benedict
    0.83
    otle
    0.80
    angelo
    0.80
    ervative
    0.78
    orio
    0.77
    Activations Density 0.005%

    No Known Activations