© 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. 1-RES-JB
    11. 3966
    Prev
    Next
    INDEX
    Explanations

    email users followed by notifications or actions

    oai_token-act-pair · gpt-3.5-turbo

    occurrences of the word "user."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 1-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    1
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Hurricanes
    -0.80
     Scand
    -0.80
     Monstrous
    -0.76
     Sacrament
    -0.74
     Baptist
    -0.73
     Patriarch
    -0.72
    hovah
    -0.72
     Kitchen
    -0.70
     Wolver
    -0.68
     Magn
    -0.68
    POSITIVE LOGITS
     interface
    0.94
     user
    0.91
     interfaces
    0.89
    interface
    0.86
     users
    0.83
     interaction
    0.78
     upgr
    0.77
    izen
    0.77
    base
    0.75
    username
    0.73
    Activations Density 0.014%

    No Known Activations