© 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 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. 8-RES-JB
    11. 1071
    Prev
    Next
    INDEX
    Explanations

    instructions related to preheating ovens

    oai_token-act-pair · gpt-3.5-turbo

    cooking instructions related to heating and baking processes

    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
    detail
    -0.89
    Interstitial
    -0.84
     pict
    -0.72
     glimps
    -0.72
    ighth
    -0.71
     vest
    -0.68
     implying
    -0.66
    details
    -0.66
    abilia
    -0.63
     Cele
    -0.63
    POSITIVE LOGITS
     oven
    0.85
     Prepare
    0.82
    Prep
    0.81
    Ingredients
    0.78
     skillet
    0.77
    ilst
    0.77
    quartered
    0.76
     prepare
    0.76
    heat
    0.75
     chilled
    0.74
    Activations Density 0.108%

    No Known Activations