© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI 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. 3460
    Prev
    Next
    INDEX
    Explanations

    information related to Roman history and archaeological sites

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot

    references to educational or informational content about history

    oai_token-act-pair · gpt-4o-miniTriggered 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
    Not in Any Lists

    No Comments

    Negative Logits
    aska
    -0.91
    Lind
    -0.91
    aukee
    -0.88
    Haw
    -0.86
     shark
    -0.85
    ket
    -0.82
     whales
    -0.82
    kef
    -0.80
     allerg
    -0.80
     Anchorage
    -0.79
    POSITIVE LOGITS
    Roman
    2.39
     Romanian
    2.34
     Romania
    2.21
     Roman
    2.19
     Buch
    1.92
     Romans
    1.81
     Rome
    1.79
     Bulgar
    1.79
     Roma
    1.70
     Bulgarian
    1.60
    Activations Density 0.279%

    No Known Activations