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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsBlogSlackPrivacy & 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. 46159
    Prev
    Next
    INDEX
    Explanations

    references to geographic locations and sports teams

    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
    430
    +0.12
    0.6%
    192
    +0.10
    0.4%
    589
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    232
    +0.12
    0.04
    359
    +0.10
    0.02
    176
    +0.09
    0.02
    Negative Logits
    DragonMagazine
    -0.84
     Donation
    -0.81
    tumblr
    -0.76
     portfolios
    -0.73
    uscript
    -0.72
    staking
    -0.72
     pseudonym
    -0.71
     portfolio
    -0.70
    minist
    -0.69
     vou
    -0.69
    POSITIVE LOGITS
     rivals
    0.94
     challenger
    0.82
     challengers
    0.81
     teammate
    0.80
     opponent
    0.80
     rival
    0.80
     opponents
    0.79
    hend
    0.74
     Olymp
    0.74
     foe
    0.72
    Activations Density 0.125%

    No Known Activations