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. 1700
    Prev
    Next
    INDEX
    Explanations

    comparisons involving quantities or degrees

    oai_token-act-pair · gpt-3.5-turbo

    comparative phrases that suggest a contrast or duality

    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
    emn
    -0.51
    iatus
    -0.49
    enger
    -0.49
     Peak
    -0.48
    ²¾
    -0.46
     Pixie
    -0.46
    ety
    -0.45
    earch
    -0.44
    aba
    -0.44
    tains
    -0.44
    POSITIVE LOGITS
    etheless
    0.62
     apologies
    0.56
     WARN
    0.52
    fortunately
    0.49
     preferably
    0.49
    entimes
    0.48
    markets
    0.46
     preferring
    0.45
     economists
    0.45
     congr
    0.45
    Activations Density 1.923%

    No Known Activations