© 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. 8-RES-JB
    11. 11844
    Prev
    Next
    INDEX
    Explanations

    phrases that introduce or summarize a statement, such as "In other words."

    oai_token-act-pair · gpt-3.5-turbo

    phrases that convey key conclusions or takeaways

    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
    )",
    -0.77
    .,"
    -0.70
    )</
    -0.66
    anus
    -0.66
    obi
    -0.66
    Scroll
    -0.65
     recalls
    -0.64
    Newsletter
    -0.64
    igree
    -0.64
    ");
    -0.63
    POSITIVE LOGITS
     ALWAYS
    0.79
     Canaver
    0.78
     fucked
    0.77
     goddamn
    0.77
     NOT
    0.76
     REALLY
    0.75
     genuinely
    0.75
     shitty
    0.74
     objectively
    0.74
     absolutely
    0.73
    Activations Density 1.499%

    No Known Activations