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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI 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. 11-RES-JB
    11. 5582
    Prev
    Next
    INDEX
    Explanations

    phrases related to political or governmental topics, especially questioning actions or decisions

    oai_token-act-pair · gpt-3.5-turbo

    phrases related to criticisms of policies or actions, especially in a political context

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 11-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    11
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Brunswick
    -0.67
     Bloom
    -0.66
     Drawn
    -0.66
     Glou
    -0.65
     Uncharted
    -0.63
     Rhodes
    -0.63
     Syracuse
    -0.62
     Scarlet
    -0.62
     Lancaster
    -0.61
     Bliss
    -0.61
    POSITIVE LOGITS
    should
    1.16
    shall
    0.99
    needs
    0.94
    vernment
    0.91
    would
    0.90
    ufact
    0.88
    selves
    0.88
    existence
    0.88
    poses
    0.88
    might
    0.87
    Activations Density 0.230%

    No Known Activations