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

    Neuronpedia

    Jacobian LensNEW
    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. 1-RES-JB
    11. 21651
    Prev
    Next
    INDEX
    Explanations

    the word "do" in sentences where someone is being compelled or urged to take action

    oai_token-act-pair · gpt-3.5-turbo

    instances of the phrase "do" followed by related actions or obligations

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 1-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    1
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Handling
    -0.78
    dayName
    -0.69
    Reviewer
    -0.69
    Software
    -0.67
     Hok
    -0.66
     Runes
    -0.65
     Secondly
    -0.65
    Tai
    -0.64
     Quarter
    -0.62
    lights
    -0.62
    POSITIVE LOGITS
    omsday
    1.10
    pez
    1.03
    ppel
    1.00
    lez
    0.96
    oley
    0.87
    xx
    0.83
    lyak
    0.82
    ggy
    0.81
    berman
    0.81
    oway
    0.80
    Activations Density 0.114%

    No Known Activations