© 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. 12-RES-JB
    11. 2433
    Prev
    Next
    INDEX
    Explanations

    names or terms related to specific people or entities

    oai_token-act-pair · gpt-3.5-turbo
    Triggered by @bot

    the name "Enron" and references to the Enron scandal

    oai_token-act-pair · gpt-4o-mini
    Triggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 12-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.hook_resid_post
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    Data Type
    torch.float32
    Hook Point
    blocks.11.hook_resid_post
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    11
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Negative Logits
     pause
    -0.69
     subject
    -0.67
     overdue
    -0.67
     free
    -0.65
     past
    -0.64
     follow
    -0.64
     representation
    -0.63
     aliens
    -0.61
     appropriate
    -0.60
     stabilization
    -0.60
    POSITIVE LOGITS
    cyclop
    1.44
    chanted
    1.42
    semble
    1.35
    voy
    1.28
    riched
    1.27
    viron
    1.26
    igma
    1.26
    igmatic
    1.24
    nis
    1.22
    vironments
    1.21
    Activations Density 0.009%

    No Known Activations