© 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. 3-RES-JB
    11. 24203
    Prev
    Next
    INDEX
    Explanations

    comparisons or contrasts indicated by the word "Whereas"

    oai_token-act-pair · gpt-3.5-turbo

    the word "whereas," indicating comparisons or contrasts between two subjects

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    3
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    hal
    -0.86
    aer
    -0.85
    entry
    -0.81
    hor
    -0.70
    Lean
    -0.70
    ve
    -0.69
    bard
    -0.69
    Hack
    -0.68
    iola
    -0.68
    omo
    -0.68
    POSITIVE LOGITS
    xual
    1.11
    lihood
    0.87
    soever
    0.85
    BILITIES
    0.80
    entimes
    0.76
    wcsstore
    0.71
     Whereas
    0.71
    sembly
    0.70
     CLIENT
    0.69
    tons
    0.67
    Activations Density 0.007%

    No Known Activations