© 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. 22354
    Prev
    Next
    INDEX
    Explanations

    phrases related to politics, specifically mentions of the Democratic Party

    oai_token-act-pair · gpt-3.5-turbo

    references to the word "Democrat" and its variations in various contexts

    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
     knots
    -0.79
     compound
    -0.71
    pell
    -0.70
     OCD
    -0.70
     processed
    -0.69
     Coat
    -0.68
     Tau
    -0.68
     knot
    -0.68
     Kathryn
    -0.67
     residue
    -0.67
    POSITIVE LOGITS
    Dem
    3.14
    dem
    2.74
    DEM
    2.15
     DEM
    1.87
     Demo
    1.32
    democracy
    1.30
    democratic
    1.27
     democrat
    1.27
     dem
    1.24
     Dem
    1.18
    Activations Density 0.021%

    No Known Activations