© 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. 24239
    Prev
    Next
    INDEX
    Explanations

    the word "dig" at different activation levels

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "dig" and its variations, indicating excavation or investigation contexts

    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
     Leilan
    -0.66
    âĹ¼
    -0.65
    BuyableInstoreAndOnline
    -0.64
    ÙĦ
    -0.59
     Karin
    -0.57
    Number
    -0.56
     Watt
    -0.56
     Mara
    -0.55
    Male
    -0.55
    Anderson
    -0.55
    POSITIVE LOGITS
    gers
    1.07
    ging
    1.06
     trenches
    1.06
    ression
    1.05
    itial
    1.04
     dug
    1.01
    raph
    1.01
    ress
    0.98
    ested
    0.96
     dig
    0.93
    Activations Density 0.034%

    No Known Activations