© 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. 7-RES-JB
    11. 23088
    Prev
    Next
    INDEX
    Explanations

    phrases starting with "Well,"

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "well" as a discourse marker

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 7-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    7
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    İĭ
    -0.70
    FN
    -0.67
    mage
    -0.63
    untarily
    -0.62
    =~=~
    -0.60
     neighb
    -0.60
    coni
    -0.60
    emo
    -0.60
    sk
    -0.59
    atile
    -0.59
    POSITIVE LOGITS
     yeah
    1.05
     uh
    0.94
     guess
    0.92
     maybe
    0.90
     fortunately
    0.87
     congr
    0.87
     luckily
    0.87
     congratulations
    0.84
     yes
    0.84
     sorry
    0.83
    Activations Density 0.047%

    No Known Activations