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. 21898
    Prev
    Next
    INDEX
    Explanations

    names that sound like "Klein" or "Zeppelin"

    oai_token-act-pair · gpt-3.5-turbo

    the repeated mention of specific individuals or terms associated with academic or research 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
    RAW
    -0.69
    ements
    -0.68
     sund
    -0.66
    ATIONAL
    -0.65
     Af
    -0.62
    sav
    -0.62
     dehyd
    -0.62
    ARE
    -0.60
     herpes
    -0.59
     Abe
    -0.59
    POSITIVE LOGITS
    lein
    3.57
    ppelin
    2.77
    pez
    1.92
    ppo
    1.31
    ppel
    1.26
    geist
    1.18
    infeld
    1.10
    etheus
    1.09
    steen
    1.07
    kus
    1.04
    Activations Density 0.061%

    No Known Activations