© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW 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. 20507
    Prev
    Next
    INDEX
    Explanations

    phrases related to human interaction

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "interact" and related forms

    oai_token-act-pair · gpt-4o-mini
    Triggered 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
    Negative Logits
    cision
    -0.72
    ft
    -0.71
    prus
    -0.67
    aft
    -0.66
    enthal
    -0.65
    FC
    -0.64
    fc
    -0.64
    statement
    -0.64
    nic
    -0.63
    secution
    -0.62
    POSITIVE LOGITS
    ivity
    1.08
     interact
    1.06
    uate
    1.00
     interacts
    0.99
     interactions
    0.99
     interacted
    0.99
     interacting
    0.93
     interaction
    0.90
    ively
    0.86
    acebook
    0.86
    Activations Density 0.011%

    No Known Activations