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. 10647
    Prev
    Next
    INDEX
    Explanations

    descriptive words related to cooking methods, such as "roasted" and "grilled"

    oai_token-act-pair · gpt-3.5-turbo

    cooking methods and techniques related to roasting and grilling food

    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
    bay
    -0.86
     Habit
    -0.79
     Oliv
    -0.73
     Facilities
    -0.72
     Bonds
    -0.70
    ynes
    -0.69
     Wak
    -0.69
    eworks
    -0.67
    idem
    -0.67
    icide
    -0.67
    POSITIVE LOGITS
     roasted
    2.36
     roast
    2.25
     grilled
    2.14
     branded
    1.36
     grill
    1.27
     lamb
    1.13
    cision
    1.11
     nailed
    1.10
     cruc
    1.02
    usky
    1.00
    Activations Density 0.053%

    No Known Activations