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. 3-RES-JB
    6. 1241
    Prev
    Next
    INDEX
    Explanations

    the phrase "sum of" followed by a number

    oai_token-act-pair · gpt-3.5-turbo

    the concept of "sum" in various mathematical and figurative contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    3
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     surn
    -0.78
     blinking
    -0.69
     foreclosure
    -0.69
     Kers
    -0.65
     Defenders
    -0.63
     tradem
    -0.63
     Ferry
    -0.63
     clipboard
    -0.62
     livest
    -0.60
    âĹ¼
    -0.60
    POSITIVE LOGITS
    ming
    1.23
    mits
    1.12
     sum
    0.95
    mers
    0.91
    sum
    0.87
    atra
    0.86
    mes
    0.85
     sums
    0.84
    otaur
    0.78
    unity
    0.77
    Activations Density 0.007%

    No Known Activations