© 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. 22370
    Prev
    Next
    INDEX
    Explanations

    instances where the directive to "consider" is used, often followed by a topic or proposition

    oai_token-act-pair · gpt-3.5-turbo

    phrases that prompt the reader to consider or analyze specific ideas or examples

    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
    "}],"
    -0.76
    "},"
    -0.72
    ially
    -0.70
    eal
    -0.64
    Contents
    -0.63
    etermined
    -0.62
    icipated
    -0.62
    dar
    -0.62
    ][/
    -0.60
    IAL
    -0.60
    POSITIVE LOGITS
    able
    0.77
     yourselves
    0.76
    tainment
    0.73
     Exhibit
    0.71
    amaz
    0.71
     examples
    0.71
    andum
    0.68
     Compare
    0.68
    asm
    0.67
     imagine
    0.65
    Activations Density 0.095%

    No Known Activations