© 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. 1945
    Prev
    Next
    INDEX
    Explanations

    terms related to the action of reconstructing or rebuilding

    oai_token-act-pair · gpt-3.5-turbo

    terms related to reconstruction processes

    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
    cius
    -0.90
    nit
    -0.79
    die
    -0.73
    Demand
    -0.73
    dm
    -0.71
    Nav
    -0.71
    danger
    -0.70
    hibit
    -0.70
    cca
    -0.70
    ILA
    -0.69
    POSITIVE LOGITS
     reconstruction
    1.48
     reconstruct
    1.24
     reconstructed
    1.17
     Reconstruction
    1.13
     rebuilt
    0.85
     pse
    0.81
     excav
    0.79
     decrypt
    0.78
     gobl
    0.77
     recre
    0.76
    Activations Density 0.007%

    No Known Activations