© 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. 0-RES-JB
    11. 9631
    Prev
    Next
    INDEX
    Explanations

    numerical values in the thousands

    oai_token-act-pair · gpt-3.5-turbo

    occurrences of large numerical values, particularly in the thousands

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     ben
    -0.72
     reflection
    -0.60
     Beat
    -0.59
     matchup
    -0.58
     favored
    -0.58
     matchups
    -0.57
     glor
    -0.57
    kell
    -0.57
     Gerald
    -0.56
     just
    -0.56
    POSITIVE LOGITS
    000
    3.75
    0002
    2.35
    0000
    2.13
    0001
    2.03
    00000
    1.97
    00
    1.94
    001
    1.86
    500
    1.83
    0010
    1.75
     000
    1.73
    Activations Density 0.056%

    No Known Activations