© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    5. GPT2-Small
    7. Residual Stream
    9. 3-RES-JB
    11. 5846
    Prev
    Next
    INDEX
    Explanations

    mentions of colors or colours

    oai_token-act-pair · gpt-3.5-turbo

    references to colors

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    3
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    idem
    -0.84
    doms
    -0.79
    _-
    -0.78
    TI
    -0.76
    ren
    -0.70
    ctors
    -0.70
    amen
    -0.68
    Xi
    -0.68
    York
    -0.67
    ement
    -0.67
    POSITIVE LOGITS
     colors
    1.31
     colours
    1.25
     palette
    1.10
    pace
    1.03
     dots
    1.02
    creen
    0.99
     color
    0.99
    anguage
    0.97
     stripes
    0.97
     tint
    0.95
    Activations Density 0.008%

    No Known Activations