© 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. 4-RES-JB
    11. 7271
    Prev
    Next
    INDEX
    Explanations

    words related to steel, such as "Steel", "Electric Light Orchestra", and "Flying Car"

    oai_token-act-pair · gpt-3.5-turbo

    references to specific bands and music-related terms

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

    No Comments

    Negative Logits
     dwar
    -0.73
     mc
    -0.72
     coron
    -0.70
    usc
    -0.70
     clipboard
    -0.69
     equ
    -0.67
    choes
    -0.65
    aciously
    -0.65
    etsy
    -0.64
    ullivan
    -0.63
    POSITIVE LOGITS
     Dogs
    1.08
     Feet
    1.02
     Riders
    0.99
     Dreams
    0.99
     Club
    0.99
     Zone
    0.98
     Games
    0.97
     Fighters
    0.96
     Track
    0.95
     Strip
    0.94
    Activations Density 0.058%

    No Known Activations