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. 10-RES-JB
    6. 2033
    Prev
    Next
    INDEX
    Explanations

    descriptions of food ingredients and cooking methods

    oai_token-act-pair · gpt-3.5-turbo

    descriptions of food textures and flavors

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    10
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    agents
    -0.83
    Ops
    -0.71
    dule
    -0.71
     Robots
    -0.71
     Ethics
    -0.70
    Enter
    -0.69
    source
    -0.69
     indoctr
    -0.69
     Destination
    -0.69
     Rent
    -0.69
    POSITIVE LOGITS
     creamy
    1.44
     matte
    1.39
     smooth
    1.37
     softer
    1.32
     shimmer
    1.30
     crisp
    1.30
     smoother
    1.28
     thicker
    1.23
     glossy
    1.19
     darker
    1.19
    Activations Density 0.344%

    No Known Activations