Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 11-RES-JB
    6. 19770
    Prev
    Next
    INDEX
    Explanations

    mentions of political asylum seekers and their associated countries

    oai_token-act-pair · gpt-3.5-turbo

    references to political refugees and their circumstances

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 11-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    11
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     unison
    -0.88
     Chevy
    -0.78
     Quarter
    -0.76
     halftime
    -0.76
     Bowl
    -0.76
     Darrell
    -0.75
    geist
    -0.74
    bowl
    -0.72
     Playoff
    -0.72
     Craft
    -0.71
    POSITIVE LOGITS
     deported
    1.44
     asylum
    1.43
     exile
    1.40
     repatri
    1.38
     extradition
    1.38
     deportation
    1.36
     extrad
    1.34
     smuggled
    1.31
     smugg
    1.28
     refugee
    1.26
    Activations Density 0.591%

    No Known Activations