© 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. 8-RES-JB
    11. 17825
    Prev
    Next
    INDEX
    Explanations

    phrases rooted in the word "Tab" followed by a number

    oai_token-act-pair · gpt-3.5-turbo

    references to the word "Tab" or variations of it

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    IVERS
    -0.75
    女
    -0.69
    IGH
    -0.68
    ĨĴ
    -0.68
    Downloadha
    -0.67
    ngth
    -0.65
     CHRIST
    -0.63
    isner
    -0.62
     Flavoring
    -0.62
    lihood
    -0.62
    POSITIVE LOGITS
    bed
    1.23
    ular
    1.22
    ulated
    1.21
    bing
    1.13
    lished
    1.13
    ulation
    1.10
    oola
    1.09
    ulator
    1.03
    ulations
    1.03
    ula
    0.99
    Activations Density 0.041%

    No Known Activations