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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & 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. 3-RES-JB
    6. 2039
    Prev
    Next
    INDEX
    Explanations

    references to different types of hair color, particularly blonde hair

    oai_token-act-pair · gpt-3.5-turbo

    references to hair color, specifically blonde

    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
    ablishment
    -0.89
    ourke
    -0.83
    apego
    -0.83
    inventoryQuantity
    -0.82
    displayText
    -0.81
    GAN
    -0.77
    arnaev
    -0.75
    ribly
    -0.74
    arcity
    -0.74
    arters
    -0.73
    POSITIVE LOGITS
     wig
    1.25
     blond
    1.20
    haired
    1.13
     blonde
    1.10
    bread
    1.06
     hair
    1.01
    beard
    0.98
    grass
    0.96
     curly
    0.94
     bombshell
    0.91
    Activations Density 0.009%

    No Known Activations