© 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. 2-RES-JB
    11. 19028
    Prev
    Next
    INDEX
    Explanations

    phrases where someone is speaking or quoting, especially with the word "said" followed by another verb

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "said" and its variations, indicating speech or quotation

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 2-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    2
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    PHOTOS
    -0.84
    ntil
    -0.79
    Newsletter
    -0.79
    uras
    -0.77
    Tai
    -0.76
    Jer
    -0.74
    Tele
    -0.74
    =~=~
    -0.73
    âĹı
    -0.71
    pleting
    -0.70
    POSITIVE LOGITS
     goodbye
    1.15
     hello
    0.75
     aloud
    0.75
     Goodbye
    0.73
     sarcast
    0.72
     afterward
    0.70
     bluntly
    0.68
     afterwards
    0.68
     anecd
    0.66
     Sek
    0.66
    Activations Density 0.287%

    No Known Activations