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

    Neuronpedia

    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. 4803
    Prev
    Next
    INDEX
    Explanations

    specific mentions of a well-known coffee company

    oai_token-act-pair · gpt-3.5-turbo

    references to coffee-related terms and products

    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
    Yan
    -0.85
    olver
    -0.81
    Metal
    -0.79
     Snake
    -0.76
     Yan
    -0.74
     Metal
    -0.74
    NRS
    -0.73
     snake
    -0.73
     WWE
    -0.71
     snakes
    -0.70
    POSITIVE LOGITS
    ffe
    3.97
    ffee
    2.17
     Starbucks
    1.73
     Fra
    1.68
    resso
    1.54
     espresso
    1.52
     caf
    1.51
    efe
    1.50
     coff
    1.45
     cafes
    1.34
    Activations Density 0.027%

    No Known Activations