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 49152
    5. 8-RES_FS49152-JB
    6. 41373
    Prev
    Next
    INDEX
    Explanations

    references to academic studies and research

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs49152-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    49,152
    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
    724
    +0.12
    0.5%
    447
    +0.11
    0.5%
    114
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    114
    +0.12
    0.03
    419
    +0.11
    0.02
    345
    +0.11
    0.02
    Negative Logits
     Rapids
    -0.68
     Pole
    -0.63
     Fiesta
    -0.62
    roots
    -0.61
    riot
    -0.60
     groceries
    -0.60
     lookout
    -0.60
    cakes
    -0.59
    gets
    -0.58
     airst
    -0.58
    POSITIVE LOGITS
     conducted
    1.15
     involving
    1.03
     performed
    1.02
     studies
    0.94
     examining
    0.94
    uggest
    0.94
     undertaken
    0.93
     published
    0.90
     evaluating
    0.90
     comparing
    0.88
    Activations Density 0.033%

    No Known Activations