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. 10-RES-JB
    6. 15600
    Prev
    Next
    INDEX
    Explanations

    phrases indicating a pattern or example that could be followed or imitated for improvement or as a model

    oai_token-act-pair · gpt-3.5-turbo

    instances of the phrase "for" followed by various contexts that imply modeling or serving as an example

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    10
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    FW
    -0.74
    books
    -0.70
    aeus
    -0.69
    tw
    -0.69
    book
    -0.69
    hya
    -0.66
    heart
    -0.66
    Includes
    -0.65
    fleet
    -0.62
     accounted
    -0.62
    POSITIVE LOGITS
     future
    1.19
     sorts
    1.11
     aspiring
    1.05
     subsequent
    0.94
     others
    0.92
     eventual
    0.90
     achieving
    0.87
     constructing
    0.82
     further
    0.81
     establishing
    0.81
    Activations Density 0.228%

    No Known Activations