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. 332
    Prev
    Next
    INDEX
    Explanations

    words related to a specific character or name "Saiyaman"

    oai_token-act-pair · gpt-3.5-turbo

    references to the character "Aman."

    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
     stre
    -0.65
     BMW
    -0.62
     Fle
    -0.62
     spr
    -0.62
     Index
    -0.61
     Spike
    -0.61
     redist
    -0.60
     compact
    -0.59
     regulated
    -0.59
     Count
    -0.58
    POSITIVE LOGITS
    aman
    4.73
    amen
    2.04
    amon
    1.55
    amas
    1.34
    anan
    1.34
    amar
    1.29
    adan
    1.26
    ama
    1.23
    amac
    1.20
    eman
    1.18
    Activations Density 0.012%

    No Known Activations