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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 3-RES-JB
    11. 7278
    Prev
    Next
    INDEX
    Explanations

    occurrences of the word "parties" with varying levels of relevance

    oai_token-act-pair · gpt-3.5-turbo

    mentions of groups or entities involved in negotiations or agreements

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    3
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Gard
    -0.66
     Cunningham
    -0.64
     Ridge
    -0.63
     Charleston
    -0.62
     Regiment
    -0.62
     Raleigh
    -0.61
    Trend
    -0.60
     Removal
    -0.60
     Dear
    -0.60
     Kov
    -0.59
    POSITIVE LOGITS
    hip
    1.05
     parties
    1.03
     Parties
    0.96
    ensical
    0.88
    etter
    0.83
    ratulations
    0.81
     vying
    0.80
     thereto
    0.80
    folk
    0.79
    pring
    0.78
    Activations Density 0.027%

    No Known Activations