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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI 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. 0-RES-JB
    11. 15281
    Prev
    Next
    INDEX
    Explanations

    phrases related to physical barriers or structures

    oai_token-act-pair · gpt-3.5-turbo

    terms related to barriers or protective structures, specifically "wall" in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     coordinates
    -0.70
     inactive
    -0.68
     abbrevi
    -0.68
     operators
    -0.64
     adopt
    -0.61
     Inc
    -0.59
     Sak
    -0.59
     adopted
    -0.58
     coordinate
    -0.58
    ãĤµ
    -0.58
    POSITIVE LOGITS
    wall
    4.67
    Wall
    1.79
     wall
    1.60
    wal
    1.44
     Wall
    1.34
     Walls
    1.25
    wire
    1.20
     walls
    1.15
    wark
    1.07
    wich
    1.06
    Activations Density 0.006%

    No Known Activations