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. 0-RES-JB
    6. 10610
    Prev
    Next
    INDEX
    Explanations

    references to types of fish

    oai_token-act-pair · gpt-3.5-turbo

    references to different types of fish, particularly salmon and trout

    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
     Garg
    -0.73
     Locked
    -0.67
     Bie
    -0.64
     Stub
    -0.63
    Known
    -0.62
    IB
    -0.62
     MK
    -0.62
    Play
    -0.61
    INST
    -0.61
     Cop
    -0.61
    POSITIVE LOGITS
     salmon
    3.82
     Salmon
    2.93
     trout
    2.42
     tuna
    1.88
     Trout
    1.50
     seafood
    1.37
     lobster
    1.32
     fish
    1.30
     fisheries
    1.30
     coral
    1.26
    Activations Density 0.024%

    No Known Activations