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. 18457
    Prev
    Next
    INDEX
    Explanations

    terms related to the "Dark" franchise

    oai_token-act-pair · gpt-3.5-turbo

    references to "Dark" in various contexts, likely related to media such as games or movies

    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
     duly
    -0.72
     competing
    -0.70
     AAP
    -0.69
     includ
    -0.66
    kson
    -0.64
     arresting
    -0.61
     downhill
    -0.61
     disqualified
    -0.61
     challenged
    -0.60
     bras
    -0.59
    POSITIVE LOGITS
     Horse
    1.08
     Ages
    1.02
     Souls
    0.99
     Matter
    0.99
    moon
    0.94
    steel
    0.92
     Clouds
    0.92
     Knight
    0.92
     Horses
    0.92
    est
    0.90
    Activations Density 0.019%

    No Known Activations