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. 23271
    Prev
    Next
    INDEX
    Explanations

    references to personal pronouns and forms of the verb "to be"

    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
    446
    +0.10
    0.5%
    756
    +0.10
    0.4%
    496
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    566
    +0.10
    0.07
    697
    +0.10
    0.05
    139
    +0.10
    0.05
    Negative Logits
    venge
    -0.71
    coming
    -0.64
     Bout
    -0.61
     Ampl
    -0.61
     Reverse
    -0.61
     Desc
    -0.60
     Enhanced
    -0.60
    "]=>
    -0.59
     Amen
    -0.59
     bearer
    -0.57
    POSITIVE LOGITS
     merely
    1.25
     simply
    1.03
    Instead
    0.96
     instead
    0.96
    Nevertheless
    0.95
     only
    0.93
    Nonetheless
    0.91
     spor
    0.89
     nevertheless
    0.83
     Instead
    0.81
    Activations Density 0.673%

    No Known Activations