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. 10755
    Prev
    Next
    INDEX
    Explanations

    HTML tags and programming-related keywords like 'boost', 'char' and 'URI'

    oai_token-act-pair · gpt-3.5-turbo

    less-than signs and associated symbols, typically used in markup or programming contexts

    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
     Polk
    -0.68
     clock
    -0.63
     Rodgers
    -0.61
     Peterson
    -0.60
     laps
    -0.59
     pronounce
    -0.56
     compound
    -0.55
     practices
    -0.55
     Guer
    -0.55
     spectacular
    -0.55
    POSITIVE LOGITS
    <
    3.94
     <
    2.30
    <?
    2.28
    .<
    2.21
    </
    2.04
    <<
    1.94
     "<
    1.76
    \<
    1.75
     <[
    1.69
    ><
    1.68
    Activations Density 0.014%

    No Known Activations