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. 9217
    Prev
    Next
    INDEX
    Explanations

    mentions of ducks

    oai_token-act-pair · gpt-3.5-turbo

    references to ducks in various contexts

    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
     Palestin
    -0.84
    ccording
    -0.68
    arcer
    -0.66
     Reloaded
    -0.66
    Interstitial
    -0.65
     accur
    -0.64
    Grant
    -0.63
    addon
    -0.62
    industrial
    -0.62
    iasm
    -0.61
    POSITIVE LOGITS
     duck
    1.42
     ducks
    1.32
    lings
    1.10
     Duck
    1.05
    fish
    0.94
    weed
    0.86
    lift
    0.84
    bowl
    0.84
    bird
    0.81
    buck
    0.81
    Activations Density 0.006%

    No Known Activations