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. 3-RES-JB
    6. 2746
    Prev
    Next
    INDEX
    Explanations

    words describing commonly used items or practices

    oai_token-act-pair · gpt-3.5-turbo

    references to staple food items and their importance in diets

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    3
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    odore
    -0.89
    vernment
    -0.86
    hani
    -0.78
    apego
    -0.78
    oping
    -0.70
    TPPStreamerBot
    -0.70
    irlf
    -0.68
    izons
    -0.68
    sterdam
    -0.67
    otos
    -0.66
    POSITIVE LOGITS
     staples
    1.01
     staple
    1.00
    stay
    0.90
    tenance
    0.75
    marks
    0.75
    fields
    0.73
    é¾įå¥ij士
    0.72
    weights
    0.72
    ãĥ¢
    0.71
    essential
    0.71
    Activations Density 0.009%

    No Known Activations