© 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. 10-RES-JB
    11. 1432
    Prev
    Next
    INDEX
    Explanations

    phrases with specific formatting patterns, potentially used to demarcate sections or headings in a document

    oai_token-act-pair · gpt-3.5-turbo

    numerals or sequences of numbers in the text

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    10
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     biod
    -0.74
    cius
    -0.66
     neighb
    -0.66
     extrad
    -0.65
     undermin
    -0.65
     inactive
    -0.64
     ageing
    -0.64
     outwe
    -0.64
     manuscripts
    -0.63
     reopen
    -0.62
    POSITIVE LOGITS
    ³³³³³³³³
    1.21
    ³³³³³³³³³³³³³³³³
    1.16
    ³³³³
    1.12
    ³³³
    1.11
    ³³
    1.02
    According
    1.00
    Specifically
    0.97
    SPONSORED
    0.97
    Seriously
    0.96
    Yesterday
    0.92
    Activations Density 0.412%

    No Known Activations