© 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. 0-RES-JB
    11. 19268
    Prev
    Next
    INDEX
    Explanations

    expressions indicating causality or reason

    oai_token-act-pair · gpt-3.5-turbo

    phrases indicating causation or reasons

    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
    OTO
    -0.69
    urches
    -0.66
    inav
    -0.61
    hat
    -0.60
    DP
    -0.60
    MN
    -0.55
    cknow
    -0.55
    ogle
    -0.54
    gang
    -0.54
    RY
    -0.54
    POSITIVE LOGITS
     due
    3.54
    due
    2.54
     Due
    2.33
     owing
    2.24
    Due
    1.88
     thanks
    1.50
     because
    1.25
     attributable
    1.12
    because
    1.05
     bec
    1.05
    Activations Density 0.027%

    No Known Activations