© 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. 11-RES-JB
    11. 20494
    Prev
    Next
    INDEX
    Explanations

    verbs conveying the idea of taking action or making decisions

    oai_token-act-pair · gpt-3.5-turbo

    expressions of the phrase "take" in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 11-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    11
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    nell
    -0.71
    naire
    -0.69
    tu
    -0.67
    eat
    -0.67
    ixt
    -0.66
    =-=-=-=-
    -0.64
    atum
    -0.63
    oir
    -0.63
    olini
    -0.63
    ingen
    -0.62
    POSITIVE LOGITS
     kindly
    1.36
     lightly
    1.05
     anymore
    1.00
     sides
    0.94
     any
    0.88
     advantage
    0.87
     seriously
    0.87
     anything
    0.86
     nor
    0.82
     heed
    0.81
    Activations Density 0.040%

    No Known Activations