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. 4322
    Prev
    Next
    INDEX
    Explanations

    proper nouns related to the name "Oz"

    oai_token-act-pair · gpt-3.5-turbo

    references to "Oz" and its related 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
     breast
    -0.71
     filib
    -0.68
    ivers
    -0.67
     partitions
    -0.65
     subst
    -0.65
     discrim
    -0.64
    Cla
    -0.63
    fast
    -0.63
    holders
    -0.62
    assi
    -0.61
    POSITIVE LOGITS
     Oz
    4.21
     Ãĸ
    1.41
     Iz
    1.31
     Salem
    1.22
     Coco
    1.21
     Sz
    1.18
     Os
    1.15
     Zel
    1.12
     Emerald
    1.11
     Dorothy
    1.09
    Activations Density 0.028%

    No Known Activations