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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI 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. 8-RES-JB
    11. 12497
    Prev
    Next
    INDEX
    Explanations

    ingredients related to traditional Indian cuisine, specifically ghee (clarified butter) and different spices

    oai_token-act-pair · gpt-3.5-turbo

    references to culinary ingredients and their uses in cooking or rituals

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     guiActiveUn
    -0.78
    ureau
    -0.76
    etheus
    -0.71
     srfAttach
    -0.70
     Inspector
    -0.70
    ockets
    -0.70
    Security
    -0.69
    PsyNetMessage
    -0.69
     Personnel
    -0.69
    ESA
    -0.69
    POSITIVE LOGITS
     ingredients
    1.46
     foods
    1.42
     herbal
    1.41
     herbs
    1.40
     flav
    1.36
     oils
    1.36
     beverage
    1.36
     fermented
    1.35
     beverages
    1.33
     additives
    1.29
    Activations Density 0.917%

    No Known Activations