Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & 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. 11-RES-JB
    6. 23331
    Prev
    Next
    INDEX
    Explanations

    types or categories of different items or concepts

    oai_token-act-pair · gpt-3.5-turbo

    references to groupings or categorizations

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 11-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    11
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    brance
    -0.88
    ortium
    -0.65
    potion
    -0.64
    $$
    -0.64
    abuse
    -0.63
    hire
    -0.63
    haw
    -0.63
    displayText
    -0.62
    atown
    -0.62
    luck
    -0.61
    POSITIVE LOGITS
     phases
    1.34
     main
    1.34
     kinds
    1.30
     types
    1.29
     distinct
    1.28
     tiers
    1.21
     categories
    1.20
     pillars
    1.19
     modes
    1.16
     stages
    1.11
    Activations Density 0.141%

    No Known Activations