© 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. 0-RES-JB
    11. 5000
    Prev
    Next
    INDEX
    Explanations

    proper nouns of various geographic locations, specifically islands

    oai_token-act-pair · gpt-3.5-turbo

    references to specific islands and their relevance in various 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
     comprom
    -0.79
    cher
    -0.65
     prol
    -0.62
     rap
    -0.62
     mechan
    -0.60
     crawl
    -0.60
    acher
    -0.59
    dy
    -0.58
     jan
    -0.58
    chery
    -0.57
    POSITIVE LOGITS
     Islands
    4.43
     islands
    2.34
     Isles
    2.05
     Island
    1.95
     Islanders
    1.57
     island
    1.56
     Isle
    1.47
    agos
    1.30
     Mountains
    1.29
     reefs
    1.28
    Activations Density 0.017%

    No Known Activations