© 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. 20484
    Prev
    Next
    INDEX
    Explanations

    references to digging or excavating

    oai_token-act-pair · gpt-3.5-turbo

    references to the action of excavation or uncovering

    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
    ulf
    -0.77
    ë
    -0.75
     transient
    -0.69
    ISA
    -0.68
     Visa
    -0.67
    iah
    -0.65
    OPA
    -0.65
     variance
    -0.63
    isa
    -0.62
     conformity
    -0.61
    POSITIVE LOGITS
     dug
    3.83
     digging
    2.11
     digs
    1.94
     dig
    1.89
     unearthed
    1.68
     excav
    1.63
     excavation
    1.35
     trenches
    1.35
     drilled
    1.30
     mined
    1.22
    Activations Density 0.018%

    No Known Activations