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. 14504
    Prev
    Next
    INDEX
    Explanations

    references to the game Super Smash Bros

    oai_token-act-pair · gpt-3.5-turbo

    references to the "Bros" brand, particularly "Warner Bros" and variations associated with it

    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
    ICLE
    -0.71
    ULE
    -0.69
     policy
    -0.65
     short
    -0.64
    ILE
    -0.63
     fasting
    -0.63
     grants
    -0.62
     kilomet
    -0.59
     fingerprint
    -0.59
     grant
    -0.59
    POSITIVE LOGITS
     Bros
    4.06
     Brothers
    2.26
     Brother
    1.11
    Brother
    1.08
     brothers
    1.04
     Bro
    0.97
     Brav
    0.95
     Roc
    0.93
     Bender
    0.93
     Sisters
    0.93
    Activations Density 0.011%

    No Known Activations