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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 1-RES-JB
    11. 18826
    Prev
    Next
    INDEX
    Explanations

    references to locations or places named "Santa"

    oai_token-act-pair · gpt-3.5-turbo

    references to specific locations or events associated with "Santa."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 1-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    1
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    edly
    -0.68
    swing
    -0.63
    orph
    -0.63
    draw
    -0.62
    td
    -0.62
    ochond
    -0.61
    intent
    -0.60
     clipboard
    -0.60
    tips
    -0.60
    Edited
    -0.60
    POSITIVE LOGITS
     Claus
    1.56
     Santa
    1.26
    Santa
    1.14
     Clara
    1.08
     Monica
    0.94
     mosqu
    0.87
     Ana
    0.86
     Rudolph
    0.82
     Maria
    0.82
     Cruz
    0.82
    Activations Density 0.008%

    No Known Activations