© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    5. GPT2-Small
    7. Residual Stream
    9. 11-RES-JB
    11. 1996
    Prev
    Next
    INDEX
    Explanations

    phrases related to programs, aid initiatives, and support services

    oai_token-act-pair · gpt-3.5-turbo

    phrases that include the word "for" in various contexts, particularly related to support, resources, or provisions

    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
    edin
    -0.91
    é¾
    -0.83
    orbit
    -0.75
    mare
    -0.73
    alan
    -0.72
    bell
    -0.71
    auntlet
    -0.71
    hew
    -0.71
    smanship
    -0.71
    aan
    -0.70
    POSITIVE LOGITS
    bidden
    1.25
    geries
    1.16
    gery
    1.10
     detecting
    1.03
     sale
    0.98
     purposes
    0.97
     storing
    0.97
     determining
    0.95
     starters
    0.94
     beginners
    0.94
    Activations Density 0.278%

    No Known Activations