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. 4-RES-JB
    6. 8949
    Prev
    Next
    INDEX
    Explanations

    sentences ending with a full stop

    oai_token-act-pair · gpt-3.5-turbo

    punctuations, specifically periods at the end of statements

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

    No Comments

    Negative Logits
    uly
    -0.87
    iber
    -0.74
    thal
    -0.71
    arov
    -0.71
     opponent
    -0.71
    ãĤ§
    -0.71
     helper
    -0.69
     objective
    -0.68
     organis
    -0.68
     content
    -0.65
    POSITIVE LOGITS
     Worse
    1.04
     But
    1.02
     Eventually
    0.98
     Their
    0.97
     And
    0.96
     They
    0.96
     Luckily
    0.96
     Whatever
    0.94
     Amid
    0.94
     Suddenly
    0.94
    Activations Density 1.179%

    No Known Activations