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. 8-RES-JB
    6. 15482
    Prev
    Next
    INDEX
    Explanations

    phrases related to being next to a body of water

    oai_token-act-pair · gpt-3.5-turbo

    references to "side" and "ide" concepts related to coastal or seaside environments

    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
    ional
    -0.93
    gins
    -0.92
    anson
    -0.87
    icia
    -0.86
    iments
    -0.85
    orius
    -0.85
    oras
    -0.84
    lio
    -0.84
    \\\\\\\\
    -0.83
    ivia
    -0.83
    POSITIVE LOGITS
     attractions
    0.78
     hospitality
    0.76
     kios
    0.74
     terr
    0.73
     attraction
    0.73
     peninsula
    0.73
     amen
    0.73
     Spa
    0.70
     park
    0.70
     neighbour
    0.70
    Activations Density 0.132%

    No Known Activations