© 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. 0-RES-JB
    11. 23231
    Prev
    Next
    INDEX
    Explanations

    hyperlinks prompting user interaction

    oai_token-act-pair · gpt-3.5-turbo

    repeated instances of the word "click."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    landers
    -0.71
    ¶ħ
    -0.68
    bors
    -0.67
    atron
    -0.66
    ald
    -0.65
    ibi
    -0.65
    ollen
    -0.63
     retiring
    -0.62
    endi
    -0.62
    ihar
    -0.62
    POSITIVE LOGITS
     click
    3.59
    click
    2.91
     clicks
    2.69
     clicking
    2.49
     clicked
    2.37
     CLICK
    2.31
     Click
    2.24
    Click
    2.03
     scroll
    1.42
     swipe
    1.28
    Activations Density 0.012%

    No Known Activations