© 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. 14309
    Prev
    Next
    INDEX
    Explanations

    adjectives or verbs related to confusion and bewilderment

    oai_token-act-pair · gpt-3.5-turbo

    expressions of confusion and strong emotional responses

    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
     faire
    -0.67
     [+
    -0.66
     deviations
    -0.65
     equity
    -0.64
    Broad
    -0.64
     settlement
    -0.64
     reconciliation
    -0.62
     Equity
    -0.61
     appropriation
    -0.60
     bearer
    -0.58
    POSITIVE LOGITS
    ingly
    1.27
    ibly
    0.90
    vic
    0.90
    ulous
    0.87
    ishment
    0.86
    iously
    0.85
    ruciating
    0.83
    ienced
    0.83
    iven
    0.83
    ciating
    0.82
    Activations Density 0.093%

    No Known Activations