© 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. 5184
    Prev
    Next
    INDEX
    Explanations

    Roman and Greek historical figures or references, particularly focusing on names ending in 'ius'

    oai_token-act-pair · gpt-3.5-turbo

    Latin names or terms related to individuals, particularly historical or religious figures

    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
    ishers
    -0.93
    sheet
    -0.81
    ished
    -0.77
    iage
    -0.76
    owler
    -0.74
    iners
    -0.73
    ishing
    -0.72
    ning
    -0.72
    orney
    -0.71
    soDeliveryDate
    -0.70
    POSITIVE LOGITS
    aurus
    1.33
    '
    1.06
    henko
    1.00
     Mons
    0.96
    hip
    0.90
     Aure
    0.87
    CRIP
    0.87
    XM
    0.87
    peed
    0.86
     XII
    0.85
    Activations Density 0.145%

    No Known Activations