© 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. 3-RES-JB
    11. 583
    Prev
    Next
    INDEX
    Explanations

    terms related to architectural elements, specifically focusing on "lattice" and "chimney"

    oai_token-act-pair · gpt-3.5-turbo

    nouns and terms associated with structures and locations

    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
     neut
    -0.82
    know
    -0.70
    cedented
    -0.66
     unrestricted
    -0.65
    paces
    -0.65
     Galile
    -0.64
    okin
    -0.63
    uggest
    -0.63
    utenberg
    -0.63
    unte
    -0.61
    POSITIVE LOGITS
    hole
    0.85
    pipe
    0.82
    board
    0.81
    lain
    0.72
    OUS
    0.72
    breaker
    0.72
    puff
    0.71
     containing
    0.71
    BALL
    0.71
    belt
    0.70
    Activations Density 0.145%

    No Known Activations