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

    Neuronpedia

    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. 6941
    Prev
    Next
    INDEX
    Explanations

    phrases related to government policies and actions

    oai_token-act-pair · gpt-3.5-turbo

    phrases that include the word "of" followed by contextual mentions relating to various social or political systems

    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
    worthy
    -0.75
    orthy
    -0.75
    acious
    -0.74
     reperto
    -0.69
     Extras
    -0.68
    unit
    -0.66
    oor
    -0.64
    èĢħ
    -0.64
    paces
    -0.63
    bucks
    -0.63
    POSITIVE LOGITS
     ours
    0.89
     separation
    0.88
     treating
    0.85
     theirs
    0.84
     separating
    0.84
     distributing
    0.83
     relying
    0.82
     avoiding
    0.82
     hers
    0.82
     electing
    0.82
    Activations Density 0.145%

    No Known Activations