© 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
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 11-RES-JB
    6. 16342
    Prev
    Next
    INDEX
    Explanations

    people asking questions or making requests

    oai_token-act-pair · gpt-3.5-turbo

    instances of questioning or asking

    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
    fit
    -0.66
    edition
    -0.59
    amation
    -0.58
    vantage
    -0.57
    jam
    -0.57
    untarily
    -0.57
    imet
    -0.56
    rite
    -0.56
    raviolet
    -0.54
     earthqu
    -0.54
    POSITIVE LOGITS
     questions
    1.30
     why
    1.04
     QUEST
    0.97
     Questions
    0.95
     whether
    0.91
     forgiveness
    0.91
     WHY
    0.89
     question
    0.88
     permission
    0.87
    why
    0.85
    Activations Density 0.066%

    No Known Activations