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. 10399
    Prev
    Next
    INDEX
    Explanations

    references to trade relationships with the EU and related implications

    oai_token-act-pair · gpt-3.5-turbo

    references to economic and export-related metrics involving the EU

    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
    colored
    -0.84
    é¾įåĸļ士
    -0.81
     colorful
    -0.78
    behavior
    -0.78
     humor
    -0.77
    VERTISEMENT
    -0.76
     demeanor
    -0.71
    odiac
    -0.71
    eah
    -0.69
    Poké
    -0.68
    POSITIVE LOGITS
     Nato
    1.38
     recognise
    1.37
     realise
    1.35
     UKIP
    1.33
     Ukip
    1.24
     privat
    1.21
     £
    1.21
    Labour
    1.21
     Labour
    1.20
    Britain
    1.20
    Activations Density 0.825%

    No Known Activations