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. 23444
    Prev
    Next
    INDEX
    Explanations

    terms related to gaming mechanics and items

    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
    430
    +0.15
    0.7%
    232
    +0.13
    0.6%
    496
    +0.13
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    232
    +0.15
    0.05
    288
    +0.13
    0.06
    326
    +0.13
    0.05
    Negative Logits
     tremend
    -0.80
    conservancy
    -0.78
     crocod
    -0.75
     tyr
    -0.74
     hydra
    -0.73
     genetically
    -0.71
     toget
    -0.71
     dinosaur
    -0.70
     disg
    -0.70
     gastro
    -0.69
    POSITIVE LOGITS
     Issues
    1.22
     Matters
    1.17
     Products
    1.17
     Problems
    1.16
     Packs
    1.16
     Cards
    1.15
     Coverage
    1.15
     Mode
    1.15
     Events
    1.14
     Protection
    1.13
    Activations Density 0.279%

    No Known Activations