© 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. 1-RES-JB
    11. 13350
    Prev
    Next
    INDEX
    Explanations

    keywords related to political challengers

    oai_token-act-pair · gpt-3.5-turbo

    references to political or competitive opponents, specifically challengers

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 1-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    1
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    othe
    -0.72
    oho
    -0.71
    olia
    -0.70
    ocon
    -0.70
    ahon
    -0.69
    eller
    -0.69
    ohan
    -0.67
    enna
    -0.66
    gas
    -0.65
    orted
    -0.64
    POSITIVE LOGITS
     challenger
    1.44
     challengers
    1.34
     competitor
    0.92
     incumbent
    0.86
    interstitial
    0.85
     contender
    0.84
     opponent
    0.84
    chall
    0.84
    nces
    0.79
     opponents
    0.79
    Activations Density 0.005%

    No Known Activations