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 3072
    5. 8-RES_FS3072-JB
    6. 2950
    Prev
    Next
    INDEX
    Explanations

    information regarding specific initiatives, projects, or stories described in the text

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs3072-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_3072
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    3,072
    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
    393
    +0.12
    0.5%
    608
    +0.12
    0.5%
    486
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    608
    +0.12
    0.09
    381
    +0.12
    0.07
    362
    +0.10
    0.08
    Negative Logits
    iture
    -0.85
    ENCE
    -0.74
     Trader
    -0.70
    heit
    -0.69
    livion
    -0.69
    anse
    -0.67
    brance
    -0.67
    ESSION
    -0.67
    Change
    -0.65
    nance
    -0.64
    POSITIVE LOGITS
     earliest
    1.05
     quir
    0.97
     inspir
    0.97
     busiest
    0.96
     pillars
    0.95
     favorites
    0.94
     finest
    0.90
     strongest
    0.89
     oldest
    0.88
     heaviest
    0.87
    Activations Density 0.703%

    No Known Activations