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. 5975
    Prev
    Next
    INDEX
    Explanations

    mentions of the country "Russia"

    oai_token-act-pair · gpt-3.5-turbo

    mentions of Russia and its related geopolitical context

    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
     :=
    -0.70
     beads
    -0.69
    walk
    -0.68
     Desmond
    -0.68
    ath
    -0.67
     CLR
    -0.67
     GC
    -0.66
    fill
    -0.65
     lum
    -0.65
    walker
    -0.65
    POSITIVE LOGITS
     Russia
    3.59
    Russia
    3.20
     Moscow
    2.60
     Russians
    2.53
    Russian
    2.53
     Russian
    2.48
    Moscow
    2.40
     Putin
    2.36
    Putin
    2.34
     Kremlin
    2.16
    Activations Density 0.023%

    No Known Activations