Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 4-RES-JB
    6. 15052
    Prev
    Next
    INDEX
    Explanations

    references to the car brand "Volkswagen."

    oai_token-act-pair · gpt-3.5-turbo

    references to Volkswagen vehicles

    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
    nces
    -0.90
    yip
    -0.82
    olulu
    -0.80
    xual
    -0.79
    mond
    -0.76
    thodox
    -0.76
    ttp
    -0.75
    EStream
    -0.75
     Uriel
    -0.74
    efeated
    -0.73
    POSITIVE LOGITS
     Polo
    1.00
    actory
    0.79
    wagen
    0.78
    PA
    0.76
    VW
    0.76
     beetle
    0.72
     dealership
    0.70
     Golf
    0.69
    GH
    0.69
    Cola
    0.68
    Activations Density 0.005%

    No Known Activations