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. 8098
    Prev
    Next
    INDEX
    Explanations

    mentions of the name "Chow."

    oai_token-act-pair · gpt-3.5-turbo

    references to "rave" culture or related terminology

    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
    ynamic
    -0.80
    hops
    -0.77
    odynam
    -0.73
     moons
    -0.68
    ynthesis
    -0.68
    aceous
    -0.67
    ordon
    -0.67
    umbers
    -0.67
    ocket
    -0.67
    icht
    -0.67
    POSITIVE LOGITS
    llers
    0.89
    rette
    0.86
    rer
    0.84
    rers
    0.82
    MENTS
    0.80
    rs
    0.80
    meric
    0.78
    lette
    0.77
    ller
    0.77
    rin
    0.76
    Activations Density 0.029%

    No Known Activations