© 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 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. 3-RES-JB
    11. 13653
    Prev
    Next
    INDEX
    Explanations

    patterns related to structural elements or organizations

    oai_token-act-pair · gpt-3.5-turbo

    repeated mentions of "structure" in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    3
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    yah
    -0.88
    jin
    -0.81
    ya
    -0.71
     Garland
    -0.71
    nee
    -0.67
    va
    -0.66
    hawks
    -0.65
    unes
    -0.65
    travel
    -0.65
     Noon
    -0.63
    POSITIVE LOGITS
     structure
    1.07
     stru
    0.94
     Structure
    0.90
     structures
    0.89
     arrang
    0.86
    urally
    0.85
    xual
    0.85
     structured
    0.82
     formation
    0.80
     cohesion
    0.79
    Activations Density 0.016%

    No Known Activations