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. 3651
    Prev
    Next
    INDEX
    Explanations

    mentions of the term "ivory."

    oai_token-act-pair · gpt-3.5-turbo

    references to ivory and its related issues, specifically focusing on poaching and illegal trade

    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
    Dir
    -0.83
    params
    -0.75
     Packers
    -0.72
     Subway
    -0.72
     DB
    -0.72
    Brew
    -0.70
     Dir
    -0.67
     Bowman
    -0.67
     Meta
    -0.67
    DEP
    -0.65
    POSITIVE LOGITS
     ivory
    3.40
     Ivory
    2.79
    Iv
    2.02
     Ivy
    1.32
     Iv
    1.29
     Cec
    1.16
     silk
    1.09
     parchment
    1.09
    Sil
    1.08
     platinum
    1.06
    Activations Density 0.045%

    No Known Activations