© 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. 22265
    Prev
    Next
    INDEX
    Explanations

    the word "Well" at the beginning of sentences

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "Well" used as a conversational transition

    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
    illary
    -0.80
    ategory
    -0.77
    ription
    -0.69
    icipated
    -0.69
    omore
    -0.67
    rift
    -0.67
    âĹ¼
    -0.67
     Parables
    -0.66
    cffff
    -0.66
    ulia
    -0.66
    POSITIVE LOGITS
    come
    1.06
    esley
    0.99
    ington
    0.95
    spring
    0.92
     hello
    0.79
     guess
    0.77
    ness
    0.74
     congratulations
    0.74
    INGTON
    0.73
     suited
    0.71
    Activations Density 0.027%

    No Known Activations