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 1536
    5. 8-RES_FS1536-JB
    6. 1346
    Prev
    Next
    INDEX
    Explanations

    descriptions of products and features, as well as investments and economic information

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_fs1536-jb
    Configuration
    jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_1536
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    1,536
    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
    481
    +0.28
    1.3%
    408
    +0.12
    0.5%
    714
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    393
    +0.28
    0.16
    714
    +0.12
    0.09
    384
    +0.09
    0.09
    Negative Logits
    ationally
    -0.81
    umat
    -0.77
    matter
    -0.76
    oiler
    -0.76
    itans
    -0.75
    emen
    -0.75
    rio
    -0.74
    ocrat
    -0.74
    ivil
    -0.73
    chool
    -0.73
    POSITIVE LOGITS
     assorted
    1.16
     occasional
    0.94
     requisite
    0.92
     willingness
    0.92
     penchant
    0.84
     adherence
    0.84
     accompanying
    0.83
     attendant
    0.81
     susceptibility
    0.81
     inability
    0.80
    Activations Density 3.677%

    No Known Activations