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. 20978
    Prev
    Next
    INDEX
    Explanations

    mentions of the color pink

    oai_token-act-pair · gpt-3.5-turbo

    references to the word "Pink."

    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
     dependence
    -0.70
    entry
    -0.66
    SER
    -0.65
     warfare
    -0.63
    endi
    -0.63
     stabilized
    -0.63
     vari
    -0.62
    Thor
    -0.62
    ravis
    -0.62
     clause
    -0.61
    POSITIVE LOGITS
     Pink
    3.79
    Pink
    3.04
     pink
    1.81
     Purple
    1.62
     Rainbow
    1.41
     Lime
    1.25
     Yellow
    1.16
     Lesbian
    1.13
     Orange
    1.12
     Green
    1.09
    Activations Density 0.014%

    No Known Activations