© 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. 1120
    Prev
    Next
    INDEX
    Explanations

    phrases that are highlighted as important or key points in a text excerpt

    oai_token-act-pair · gpt-3.5-turbo

    key phrases or sections related to highlights in stories

    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
     challeng
    -0.98
     independ
    -0.90
     toget
    -0.90
     ener
    -0.85
    ogly
    -0.79
    eatures
    -0.79
     territ
    -0.77
    urat
    -0.76
     appre
    -0.73
     psychiat
    -0.72
    POSITIVE LOGITS
     Authorities
    1.16
     Officials
    1.14
     Investigators
    1.14
     Researchers
    1.13
     Prosecutors
    1.12
     Protesters
    1.00
     Thousands
    0.98
     Scientists
    0.98
     Emails
    0.96
     Former
    0.95
    Activations Density 0.009%

    No Known Activations