© 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. 10-RES-JB
    11. 10269
    Prev
    Next
    INDEX
    Explanations

    phrases related to criminal activities and legal cases

    oai_token-act-pair · gpt-3.5-turbo

    instances of the article "a" followed by references to individuals involved in criminal activities or misconduct

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    10
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     exercises
    -0.76
    blocks
    -0.76
    years
    -0.75
    views
    -0.75
     indexes
    -0.75
     decks
    -0.74
    files
    -0.72
    thumbnails
    -0.72
    breaks
    -0.72
    bars
    -0.72
    POSITIVE LOGITS
     colleague
    1.45
     friend
    1.25
     woman
    1.23
     classmate
    1.22
     reporter
    1.14
     stranger
    1.13
     teammate
    1.12
     fellow
    1.12
     cowork
    1.10
     bunch
    1.10
    Activations Density 0.253%

    No Known Activations