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. 1025
    Prev
    Next
    INDEX
    Explanations

    specific mentions of the location "Oak Brook" in various contexts

    oai_token-act-pair · gpt-3.5-turbo

    mentions of "Brook" or related terms in the context of a location

    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
     sig
    -0.79
     att
    -0.72
     totem
    -0.69
    uti
    -0.68
     mask
    -0.67
     Sig
    -0.64
    uters
    -0.62
     signature
    -0.61
    ugal
    -0.61
     minion
    -0.61
    POSITIVE LOGITS
     Brook
    4.10
    Brook
    2.57
    brook
    1.78
     Blooming
    1.32
     Brooklyn
    1.18
     Grove
    1.11
     Nets
    1.07
     Brighton
    1.07
     Pond
    1.04
     Brooke
    1.04
    Activations Density 0.019%

    No Known Activations