© 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 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. 8-RES-JB
    11. 19388
    Prev
    Next
    INDEX
    Explanations

    terms related to political movements or affiliations, specifically those against specific countries or governments

    oai_token-act-pair · gpt-3.5-turbo

    references to anti-government sentiments and the entities involved in them

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     srfAttach
    -0.85
    OST
    -0.71
    osc
    -0.70
    ith
    -0.70
    iths
    -0.69
    iosis
    -0.68
    LESS
    -0.67
    oscope
    -0.66
    igers
    -0.64
    Higher
    -0.63
    POSITIVE LOGITS
     sentiment
    1.12
     sentiments
    1.05
     crus
    1.03
     rhetoric
    0.88
     campaigner
    0.87
     campaigners
    0.86
     stance
    0.86
     graffiti
    0.85
     bigotry
    0.81
     initiatives
    0.81
    Activations Density 0.102%

    No Known Activations