Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 1-RES-JB
    6. 23479
    Prev
    Next
    INDEX
    Explanations

    words related to the action of "ducking"

    oai_token-act-pair · gpt-3.5-turbo

    words related to evasive actions or behaviors

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 1-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    1
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ket
    -0.73
    gregation
    -0.62
    emate
    -0.61
     accompanies
    -0.60
     appreciated
    -0.60
    externalActionCode
    -0.59
     circulated
    -0.58
     circulate
    -0.58
     Fabric
    -0.57
    obook
    -0.57
    POSITIVE LOGITS
    ucking
    1.08
    ucked
    1.06
    ortun
    0.93
    ucks
    0.88
    ucker
    0.82
    oes
    0.80
     darts
    0.77
    ernel
    0.75
    weed
    0.75
     hairs
    0.74
    Activations Density 0.006%

    No Known Activations