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 · Feature Splitting for GPT2-Small
    3. GPT2-Small
    4. Residual Feature Splitting 24576
    5. 8-RES_FS24576-JB
    6. 23499
    Prev
    Next
    INDEX
    Explanations

    phrases related to buying and selling transactions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs24576-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576
    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

    Neuron Alignment
    Index
    Value
    % of L₁
    367
    +0.12
    0.6%
    244
    +0.12
    0.5%
    249
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    245
    +0.12
    0.04
    362
    +0.12
    0.04
    526
    +0.11
    -0.00
    Negative Logits
    imental
    -0.73
    mins
    -0.70
    ¯¯
    -0.70
    hered
    -0.69
    TEXTURE
    -0.68
    poral
    -0.67
    vous
    -0.66
    isel
    -0.65
    inguished
    -0.65
    orney
    -0.65
    POSITIVE LOGITS
     auctions
    1.40
     eBay
    1.38
     auction
    1.36
     supermarkets
    1.25
     cheapest
    1.24
     Auction
    1.20
     Costco
    1.19
     sellers
    1.16
     retail
    1.15
     Alibaba
    1.15
    Activations Density 0.339%

    No Known Activations