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. 10-RES-JB
    6. 3613
    Prev
    Next
    INDEX
    Explanations

    foods and ingredients mentioned in a document

    oai_token-act-pair · gpt-3.5-turbo

    references to various types of food, particularly healthy and nutrient-dense options

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    10
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    Privacy
    -0.95
    ewitness
    -0.86
    dB
    -0.85
    oustic
    -0.78
    Charges
    -0.75
     Institution
    -0.74
     Victims
    -0.73
    Interface
    -0.73
    Investigators
    -0.73
     Ages
    -0.73
    POSITIVE LOGITS
     carrots
    1.55
     potatoes
    1.55
     rice
    1.52
     beans
    1.49
     tomatoes
    1.48
     broccoli
    1.47
     vegetables
    1.47
     spinach
    1.45
     avocado
    1.44
     peas
    1.44
    Activations Density 0.157%

    No Known Activations