© 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. 7549
    Prev
    Next
    INDEX
    Explanations

    appearances of specific numerical patterns, particularly the number 36

    oai_token-act-pair · gpt-3.5-turbo

    occurrences of specific numerical values and their associated data in a structured format

    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
    olicy
    -0.79
    prises
    -0.77
    oken
    -0.76
    liga
    -0.74
    arte
    -0.74
    ItemTracker
    -0.73
    nesday
    -0.71
    atchewan
    -0.69
    reek
    -0.69
    onite
    -0.69
    POSITIVE LOGITS
    00
    0.98
    th
    0.84
    ZI
    0.72
    650
    0.71
    50
    0.71
    01
    0.68
    36
    0.68
    375
    0.67
    75
    0.66
    384
    0.65
    Activations Density 0.024%

    No Known Activations