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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 0-RES-JB
    11. 7858
    Prev
    Next
    INDEX
    Explanations

    mentions of political affiliation, specifically the word "Democrat"

    oai_token-act-pair · gpt-3.5-turbo

    references to the Democratic Party or its members

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     trainers
    -0.65
     lamps
    -0.62
     spoil
    -0.61
    ray
    -0.59
    anks
    -0.59
     Cust
    -0.59
     coh
    -0.59
     tread
    -0.56
     baths
    -0.56
     Sapphire
    -0.56
    POSITIVE LOGITS
     Democrat
    3.68
     Democrats
    2.42
     Democratic
    2.38
    Democrats
    2.25
     Republican
    2.18
    Democratic
    2.12
    Republican
    2.00
     Dems
    1.80
    Republicans
    1.78
     Republicans
    1.75
    Activations Density 0.010%

    No Known Activations