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

    Neuronpedia

    APIAssistant AxisNEWCircuit 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. 7387
    Prev
    Next
    INDEX
    Explanations

    references to political entities or military organizations

    oai_token-act-pair · gpt-3.5-turbo

    references to various "Wings" or "Wing" groups, particularly in military or aviation 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
     contam
    -0.76
     revol
    -0.75
    Scotland
    -0.70
    rates
    -0.69
     salts
    -0.66
     precip
    -0.65
     settlements
    -0.63
     insol
    -0.62
    prov
    -0.62
     aval
    -0.62
    POSITIVE LOGITS
     Wing
    4.28
    Wing
    2.60
     Wings
    1.63
     Squadron
    1.61
     wing
    1.42
    wing
    1.30
    wings
    1.25
     wings
    1.21
     Swing
    0.96
     Tail
    0.96
    Activations Density 0.008%

    No Known Activations