© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 11-RES-JB
    11. 10054
    Prev
    Next
    INDEX
    Explanations

    formal requests or instructions

    oai_token-act-pair · gpt-3.5-turbo

    requests for actions or confirmations, particularly the word "please."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 11-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    11
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    pires
    -0.61
    lite
    -0.55
    ulated
    -0.54
    onom
    -0.51
    rist
    -0.51
     Tav
    -0.50
     somew
    -0.48
    antz
    -0.47
     JUSTICE
    -0.46
    liner
    -0.46
    POSITIVE LOGITS
     Cancel
    0.81
     verify
    0.75
     login
    0.71
     refresh
    0.68
     reload
    0.68
     retake
    0.64
     subscribe
    0.64
     enter
    0.63
     try
    0.63
    pload
    0.62
    Activations Density 0.011%

    No Known Activations