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

    Neuronpedia

    APIAssistant AxisNEWCircuit 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. 24164
    Prev
    Next
    INDEX
    Explanations

    instances of the word "Gray" with high activation values

    oai_token-act-pair · gpt-3.5-turbo

    references to the name "Gray."

    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
    istas
    -0.77
     reciproc
    -0.76
    ista
    -0.71
    ilst
    -0.69
     idol
    -0.68
     superst
    -0.65
     pist
    -0.64
     authentic
    -0.63
     inst
    -0.63
    aur
    -0.62
    POSITIVE LOGITS
     Gray
    3.91
    Gray
    3.13
    gray
    1.96
     Grey
    1.90
     gray
    1.76
    Grey
    1.71
     grey
    1.54
    grey
    1.44
     Brown
    1.40
    Brown
    1.22
    Activations Density 0.017%

    No Known Activations