© 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. 4-RES-JB
    11. 22643
    Prev
    Next
    INDEX
    Explanations

    places and modes of transportation

    oai_token-act-pair · gpt-3.5-turbo

    references to transportation hubs such as train and bus stations

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

    No Comments

    Negative Logits
    RAY
    -0.78
    sonian
    -0.77
     genetically
    -0.69
    BIT
    -0.67
    morph
    -0.66
     persecuted
    -0.66
     anthrop
    -0.66
     reproduce
    -0.65
    RGB
    -0.64
     extracted
    -0.64
    POSITIVE LOGITS
     busiest
    1.14
     entrances
    1.10
     plaza
    0.95
     Plaza
    0.94
     gates
    0.91
     kios
    0.91
     parking
    0.88
     docks
    0.88
     stations
    0.87
     attendants
    0.85
    Activations Density 0.114%

    No Known Activations