© 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 EvalsExportsinterp-engineNEWAPI 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. 5782
    Prev
    Next
    INDEX
    Explanations

    references to the name "Will"

    oai_token-act-pair · gpt-3.5-turbo

    the presence of the name "Will."

    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
     special
    -0.67
     subordinate
    -0.65
     trademark
    -0.64
     account
    -0.63
     accounting
    -0.61
     Division
    -0.60
     sector
    -0.59
    igon
    -0.59
     Chronicle
    -0.59
     Department
    -0.59
    POSITIVE LOGITS
    Will
    3.26
    will
    1.56
    Should
    1.43
    Would
    1.42
    Does
    1.40
    Did
    1.32
    Can
    1.30
    Is
    1.29
    Are
    1.24
    Could
    1.23
    Activations Density 0.023%

    No Known Activations