© 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. 3-RES-JB
    11. 16999
    Prev
    Next
    INDEX
    Explanations

    instances of instructional text indicating what actions can be taken

    oai_token-act-pair · gpt-3.5-turbo

    phrasings indicating capability or permissions

    oai_token-act-pair · gpt-4o-mini
    Triggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    3
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Negative Logits
     Mant
    -0.67
    cule
    -0.63
    ception
    -0.63
     Cortex
    -0.62
    furt
    -0.61
     revision
    -0.59
     Lis
    -0.59
    burgh
    -0.59
    é¾įå¥ij士
    -0.58
     Fighter
    -0.57
    POSITIVE LOGITS
    't
    1.46
    adian
    1.11
    berra
    1.08
    NOT
    1.05
    tera
    0.99
     afford
    0.88
     easily
    0.86
    isters
    0.83
     customize
    0.81
    atell
    0.80
    Activations Density 0.130%

    No Known Activations