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. 23132
    Prev
    Next
    INDEX
    Explanations

    references to military actions and conflicts

    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
    640
    +0.28
    1.3%
    87
    +0.13
    0.6%
    64
    +0.13
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    640
    +0.28
    0.15
    94
    +0.13
    0.11
    64
    +0.13
    -0.03
    Negative Logits
    tons
    -0.64
    boro
    -0.61
    tics
    -0.60
     Trog
    -0.59
     adolesc
    -0.57
    minist
    -0.55
     Minion
    -0.55
     Ancients
    -0.55
     endeavour
    -0.55
     engagements
    -0.54
    POSITIVE LOGITS
    advertisement
    1.14
    Article
    1.10
    ³³³³³³³³
    1.05
    Advertisement
    0.99
    ³³³³
    0.98
    ³³³
    0.95
    ³³³³³³³³³³³³³³³³
    0.93
    Meanwhile
    0.86
    Still
    0.85
    posted
    0.78
    Activations Density 0.389%

    No Known Activations