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. 0-RES-JB
    6. 3994
    Prev
    Next
    INDEX
    Explanations

    names related to Russian officials

    oai_token-act-pair · gpt-3.5-turbo

    references to prominent Russian individuals, particularly politicians and diplomats

    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
     Pear
    -0.94
     Cook
    -0.79
     cider
    -0.74
    Pear
    -0.72
     Camel
    -0.70
     TC
    -0.70
     Baylor
    -0.69
    Whe
    -0.69
     Palmer
    -0.69
     Fox
    -0.68
    POSITIVE LOGITS
     Sergei
    2.88
     Dmitry
    2.83
     Sergey
    2.68
     Mikhail
    2.51
     Dmit
    2.37
     Igor
    2.27
     Ukrainians
    2.26
     Ukrainian
    2.25
    Putin
    2.21
     Nikol
    2.20
    Activations Density 0.068%

    No Known Activations