© 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. 8-RES-JB
    11. 13137
    Prev
    Next
    INDEX
    Explanations

    verbs involving the concept of existence or occurrence

    oai_token-act-pair · gpt-3.5-turbo

    instances of the phrase "seems to be" indicating speculation or uncertainty

    oai_token-act-pair · gpt-4o-mini
    Triggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Negative Logits
    aths
    -0.75
    FUL
    -0.73
    arios
    -0.70
    alez
    -0.69
    ocaust
    -0.62
    apo
    -0.61
     Emerson
    -0.61
    aft
    -0.59
     Gree
    -0.59
    DEBUG
    -0.59
    POSITIVE LOGITS
     plateau
    0.76
     stalled
    0.67
     stuck
    0.67
     circling
    0.66
     suppl
    0.66
     dividing
    0.66
    NetMessage
    0.65
     grav
    0.65
     slowing
    0.64
     phased
    0.64
    Activations Density 0.129%

    No Known Activations