© 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. 11-RES-JB
    11. 12855
    Prev
    Next
    INDEX
    Explanations

    terms related to specific professions or skills

    oai_token-act-pair · gpt-3.5-turbo

    references to gaming or competitive experiences

    oai_token-act-pair · gpt-4o-mini
    Triggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 11-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    11
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Negative Logits
    doms
    -0.88
     iPads
    -0.88
     ceilings
    -0.85
     suspensions
    -0.84
     buildings
    -0.82
     mosques
    -0.81
     assemblies
    -0.81
     embassies
    -0.78
     castles
    -0.77
     clusters
    -0.77
    POSITIVE LOGITS
     performer
    1.43
    reader
    1.43
     practitioner
    1.39
    emaker
    1.38
    handler
    1.36
    ographer
    1.36
    erer
    1.36
    ologist
    1.36
     wearer
    1.34
    builder
    1.34
    Activations Density 0.774%

    No Known Activations