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. 16053
    Prev
    Next
    INDEX
    Explanations

    the word "Sum" followed by a number in the text

    oai_token-act-pair · gpt-3.5-turbo

    occurrences of the word "Sum."

    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
     registry
    -0.72
     laser
    -0.71
     headers
    -0.69
     embr
    -0.66
     Blasio
    -0.65
    Ħ¢
    -0.65
     lasers
    -0.64
     header
    -0.64
     boobs
    -0.64
     sexist
    -0.63
    POSITIVE LOGITS
     Sum
    4.06
    Sum
    2.58
     sum
    1.86
    sum
    1.83
     SUM
    1.62
     Summon
    1.21
     Sem
    1.10
     Tem
    1.03
     Kum
    1.03
     Cum
    1.01
    Activations Density 0.008%

    No Known Activations