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. 1-RES-JB
    6. 9288
    Prev
    Next
    INDEX
    Explanations

    the word "continuing."

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "continuing."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 1-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    1
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    otropic
    -0.75
    ogy
    -0.74
    opy
    -0.73
    olog
    -0.73
     Sheikh
    -0.73
    idden
    -0.69
    athed
    -0.69
    bid
    -0.68
    odor
    -0.68
     Parables
    -0.67
    POSITIVE LOGITS
     continuing
    0.88
     continu
    0.86
     ende
    0.83
     uninterrupted
    0.77
     subsequ
    0.77
     continuation
    0.75
     appropriations
    0.75
     continued
    0.75
     conduc
    0.75
     applic
    0.74
    Activations Density 0.006%

    No Known Activations