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. 8-RES-JB
    6. 8742
    Prev
    Next
    INDEX
    Explanations

    phrases with alphanumeric characters in a specific pattern

    oai_token-act-pair · gpt-3.5-turbo

    noun phrases that are likely related to people, places, or entities

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Cho
    -0.95
     Wass
    -0.81
     Chow
    -0.74
     Commander
    -0.74
     CHO
    -0.73
     TW
    -0.73
     Aston
    -0.72
     Chosen
    -0.71
     Prest
    -0.70
     Donovan
    -0.70
    POSITIVE LOGITS
    ian
    1.50
    ia
    1.46
    ios
    1.33
    ians
    1.32
    iol
    1.28
    iac
    1.27
    io
    1.26
    ial
    1.24
    iox
    1.23
    ien
    1.23
    Activations Density 0.283%

    No Known Activations