© 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. 0-RES-JB
    11. 1930
    Prev
    Next
    INDEX
    Explanations

    fried food items

    oai_token-act-pair · gpt-3.5-turbo

    references to fried food items

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    isSpecialOrderable
    -0.74
     Veil
    -0.74
    dp
    -0.73
    Ward
    -0.72
     Shel
    -0.70
    esse
    -0.70
    doms
    -0.70
    Div
    -0.70
     Carey
    -0.69
     Dign
    -0.68
    POSITIVE LOGITS
     fried
    3.11
     boiled
    2.25
     baked
    1.88
     roasted
    1.68
     grilled
    1.58
     frying
    1.58
     crispy
    1.52
     fermented
    1.50
     mashed
    1.48
     burnt
    1.48
    Activations Density 0.019%

    No Known Activations