© 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. 9-RES-JB
    11. 10217
    Prev
    Next
    INDEX
    Explanations

    expressions related to events or actions happening in a public or open setting

    oai_token-act-pair · gpt-3.5-turbo

    phrases related to performances or actions happening in front of an audience

    oai_token-act-pair · gpt-4o-mini
    Triggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 9-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    9
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Negative Logits
    QL
    -0.72
    itars
    -0.69
    chini
    -0.68
    MER
    -0.67
    grass
    -0.66
    pring
    -0.66
    berman
    -0.64
    isure
    -0.64
    ãĥij
    -0.63
    among
    -0.62
    POSITIVE LOGITS
     cameras
    1.03
     camera
    0.86
     crowds
    0.81
     audiences
    0.80
     webcam
    0.79
     TVs
    0.78
     doors
    0.76
     microphones
    0.75
     cheering
    0.74
     roaring
    0.72
    Activations Density 0.050%

    No Known Activations